A Meta-Analysis of the Relationship Between Abstinence and Neuropsychological Functioning in Methamphetamine Use Disorder

Abstract

Background: The potential influence of methamphetamine use on neuropsychological functioning is unclear. The aim of this this meta-analysis was to investigate the relationship between abstinence and neuropsychological functioning in people with methamphetamine use disorder. Method: The systematic review protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO: CRD42018083598). Studies were eligible if they (a) included a group that identified methamphetamine as their primary substance of use, (b) comprised participants who reported a period of abstinence from methamphetamine, (c) included healthy comparison participants, (d) included outcome measures that constituted valid and reliable cognitive tests and, (e) were published in English. The search yielded effect sizes based on 1008 abstinent methamphetamine participants and 984 healthy comparison participants. Results: Findings revealed small-to-moderate effect sizes, indicating that methamphetamine participants performed somewhat below controls on learning efficiency, visual-spatial processing, comprehension knowledge, retrieval fluency, processing speed, and psychomotor speed. Three exceptions, in which performance demonstrated no group effect, were in domains of fluid reasoning, short-term working memory, and reaction and decision speed. Discussion: The current results support the hypothesis that methamphetamine use is associated with small-to-moderate cognitive sequelae that persist beyond a period of abstinence. However, we cannot determine whether methamphetamine use leads to long-term neuropsychological impairment via structural or functional brain changes, or whether preexisting deficits in neuropsychological performance and cortical integrity are vulnerability factors for methamphetamine use, or both. Taken together, the results suggest that strong statements regarding impaired cognitive functioning in abstinent methamphetamine users are premature.

Full text

A Meta-Analysis of the Relationship Between Abstinence and
Neuropsychological Functioning in Methamphetamine Use Disorder
Candice Basterfield, Robert Hester, and Stephen C. Bowden
The University of Melbourne
Background: The potential influence of methamphetamine use on neuropsychological functioning is
unclear. The aim of this this meta-analysis was to investigate the relationship between abstinence and
neuropsychological functioning in people with methamphetamine use disorder. Method: The systematic
review protocol was registered with the International Prospective Register of Systematic Reviews
(PROSPERO: CRD42018083598). Studies were eligible if they (a) included a group that identified
methamphetamine as their primary substance of use, (b) comprised participants who reported a period of
abstinence from methamphetamine, (c) included healthy comparison participants, (d) included outcome
measures that constituted valid and reliable cognitive tests and, (e) were published in English. The search
yielded effect sizes based on 1008 abstinent methamphetamine participants and 984 healthy comparison
participants. Results: Findings revealed small-to-moderate effect sizes, indicating that methamphetamine
participants performed somewhat below controls on learning efficiency, visual-spatial processing,
comprehension knowledge, retrieval fluency, processing speed, and psychomotor speed. Three excep-
tions, in which performance demonstrated no group effect, were in domains of fluid reasoning, short-term
working memory, and reaction and decision speed. Discussion: The current results support the hypoth-
esis that methamphetamine use is associated with small-to-moderate cognitive sequelae that persist
beyond a period of abstinence. However, we cannot determine whether methamphetamine use leads to
long-term neuropsychological impairment via structural or functional brain changes, or whether preex-
isting deficits in neuropsychological performance and cortical integrity are vulnerability factors for
methamphetamine use, or both. Taken together, the results suggest that strong statements regarding
impaired cognitive functioning in abstinent methamphetamine users are premature.
General Scientific Summary
This meta-analysis presents a quantitative synthesis of studies on the neuropsychological perfor-
mance of abstinent methamphetamine users. Methamphetamine users performed below healthy
controls on learning efficiency, visual-spatial processing, comprehension knowledge, retrieval flu-
ency, processing speed, and psychomotor speed. However, there were no differences between the
groups in domains of fluid reasoning, short-term working memory, and reaction and decision speed.
The findings suggest that strong statements about methamphetamine users’ cognitive functioning are
premature as undetected confounds (e.g., quantity of use, route of administration, the extent of other
drug use factors, personality traits, undiagnosed psychiatric conditions) may exert an influence on
cognition and require further research.
Keywords: abstinence, meta-analysis, methamphetamine, neuropsychological assessment
Supplemental materials: http://dx.doi.org/10.1037/neu0000552.supp
Methamphetamine, sometimes colloquially called meth,ice,
crystal,speed, or base, is a synthetic derivative of amphetamine
and contains an additional methyl group in its chemical structure.
Aconsequenceofthisstructureisincreasedcentralnervoussystem
penetration and greater potency than its parent compound amphet-
amine (Homer, Halkitis, Moeller, & Solomon, 2013).The potential for
negative effects of methamphetamine has gained increasing public
health attention (Drug Enforcement Administration, 2017). Metham-
phetamine use is widely argued to cause cognitive impairment. For
example, Nora Volkow, director of National Institute of Drug Abuse,
This article was published Online First April 25, 2019.
Candice Basterfield, Robert Hester, and Stephen C. Bowden, Melbourne
School of Psychological Sciences, The University of Melbourne.
We thank Scott Lilienfeld for his helpful comments on this manuscript
and Steven Tarlow for his statistical consultation.
Correspondence concerning this article should be addressed to Can-
dice Basterfield, Melbourne School of Psychological Sciences, The
University of Melbourne, Parkville, Melbourne VIC 3010, Australia. E-
mail: candice101101@gmail.com
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Neuropsychology
© 2019 American Psychological Association 2019, Vol. 33, No. 5, 739–753
0894-4105/19/$12.00 http://dx.doi.org/10.1037/neu0000552
739

asserted that “The consequences of methamphetamine abuse are ter-
rible for the individual—psychologically, medically, and socially.
Abusing the drug can cause memory loss, aggression, psychotic
behavior, damage to the cardiovascular system, malnutrition, and
severe dental problems” (Volkow, 2013).
Methamphetamine and the Brain
A review of the empirical literature suggests that methamphet-
amine may affect brain functioning at multiple levels of analysis:
(a) acute and chronic effects of methamphetamine use; (b) struc-
tural and functional changes in methamphetamine use; and (c)
neuropsychological functioning in abstinent methamphetamine us-
ers.
Acute and Chronic Effects of Methamphetamine Use
The direct effects of methamphetamine include increased en-
ergy, attention, and sexual drive, as well as decreased anxiety
(Newton, Kalechstein, Duran, Vansluis, & Ling, 2004). In a lab-
oratory setting, a single dose of methamphetamine can improve
attention, visuospatial perception, processing speed, and learning
and memory in individuals with limited drug experience (Hart et
al., 2008;Johnson, Ait-Daoud, & Wells, 2000). In contrast, other
studies have found no significant effects on cognitive functioning
(Hart, Ward, Haney, Foltin, & Fischman, 2001;Sevak, Stoops,
Hays, & Rush, 2009). These negative findings may be due to
several factors. For example, outside of a laboratory setting meth-
amphetamine users’ duration of use, dosage, frequency, and route
of administration may vary greatly from those prescribed in a
controlled setting. Moreover, methamphetamine purity varies within
and between geographical locations (Community Epidemiology
Work Group, National Institute of Drug Abuse, 2006). Therefore,
the potential negative side effects of methamphetamine use may be
related to one or more largely unknown substances present in
heterogeneous street methamphetamine, which may be contribut-
ing to cognitive deficits in users. Nevertheless, there is no con-
sensus regarding whether methamphetamine use variables (e.g.,
duration of use, dosage, and frequency of use) are tied to neuro-
psychological impairment (Chang et al., 2002;Cherner et al.,
2010b;Hart, Marvin, Silver, & Smith, 2012;Hoffman et al., 2006;
Johanson et al., 2006).
Chronic users of methamphetamine users typically administer
the drug in binge cycles that last for several days. Therefore, users’
cognitive functioning may be adversely affected following larger
methamphetamine doses administered in binges. Excessive meth-
amphetamine stimulation of the sympathetic nervous system dur-
ing binges appears to contribute to an increased risk for psychotic
symptoms, anxiety, and depression (London et al., 2004;Meredith,
Jaffe, Ang-Lee, & Saxon, 2005;Zweben et al., 2004).
Structural and Functional Changes in
Methamphetamine Use
The hypothesized influence of methamphetamine on cognitive
functioning relates to its established structural and functional brain
mechanisms. Methamphetamine use results in a cortical release of
monoamines, including dopamine, serotonin, and norepinephrine
(Davidson, Gow, Lee, & Ellinwood, 2001). These neurotransmit-
ters are produced in the midbrain and brainstem and project widely
to other areas (Tekin & Cummings, 2002). Methamphetamine
appears to primarily target the dopamine transporter (DAT) sys-
tem, which is responsible for the reuptake of dopamine. In addition
to regulating dopaminergic transmission, methamphetamine re-
verses the direction of dopamine, causing increased release of
dopamine into the synapse (Khoshbouei, Wang, Lechleiter, Javitch, &
Galli, 2003). Dopamine is manufactured in the substantia nigra and
the ventral tegmental area (VTA: Blumenfeld, 2010), both of which
project to subcortical and cortical regions involved in movement,
reward, attention and working memory. (Morris et al., 2016;Tekin &
Cummings, 2002). Therefore, increased production of dopamine in
the substantia nigra and VTA may produce pervasive disturbances in
cognitive and behavioral processes.
The prevailing view is that methamphetamine use can lead to
neurotoxicity in several neurotransmitter systems. This effect may
or may not be reversible (Erinoff, 1995). Evidence of neurotoxicity
comes from animal studies that show effects on the dopaminergic
and serotonergic systems (Albers & Sonsalla, 1995;Bowyer &
Holson, 1995). In rodents, methamphetamine appears to principally
damage dopaminergic and serotonergic neurons (Linder, Young, &
Groves, 1995;Pu & Vorhees, 1993). These changes lead to a
reduction in dopamine and serotonin axonal markers and striatal
dopamine transporter density as measured by position emission
topography (PET; Albers & Sonsalla, 1995;Bowyer & Holson,
1995;Villemagne et al., 1998). However, it is unclear whether the
dosages in animal studies are representative of human consump-
tion. For example, drug-naïve animals are typically administered
large doses of methamphetamine for a few days (e.g., 2.5–10
mg/kg/injection; Albers & Sonsalla, 1995). In contrast, human
methamphetamine users typically increase their drug dosage grad-
ually with more experience with the drug (Hart et al., 2012).
In humans, research using PET to measure DAT level, a marker
of dopamine cell terminals, found that abstinent methamphetamine
users showed significant reduction in DAT density in the striatum
compared with controls (McCann et al., 1998;Volkow et al., 2001;
Sekine et al., 2003), and that reduced DAT density was associated
with motor slowing and memory impairment (Volkow et al.,
2001). Studies using
1
H magnetic resonance spectroscopy have
demonstrated that chronic methamphetamine users show low lev-
els of N-acetylaspartate (NAA), a marker of neuronal health, as
well as high levels of myoinositol (suggestive of gliosis) and
choline that have been linked to neuronal damage (Dean, Groman,
Morales, & London, 2013). In abstinent methamphetamine users,
low concentrations of NAA were reduced in the basal ganglia and
frontal white matter (suggestive of neuronal damage), and there
was an increase in choline and myoinositol in the frontal cortex
(Ernst, Chang, Leonido-Yee, & Speck, 2000;Sekine et al., 2002).
Nordahl and colleagues (2002) found that abstinent methamphet-
amine users had low levels of NAA and high levels of choline in
the anterior cingulum. Furthermore, Taylor et al. (2000) also
observed low NAA in the anterior cingulum as well as low NAA
in the basal ganglia in abstinent methamphetamine users.
In addition, several MRI studies have found varying distur-
bances in cortical and subcortical brain regions. For example,
Thompson et al. (2004) found that methamphetamine use was
associated with smaller hippocampal volumes than controls. Chang
and colleagues (2005) found that methamphetamine use was as-
sociated with larger basal ganglia. In addition, Jernigan et al.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
740 BASTERFIELD, HESTER, AND BOWDEN

(2005) found that chronic methamphetamine use was associated
with abnormally large volumes in the striatum, nucleus accum-
bens, and parietal lobes.
These findings have been interpreted as consistent with neuronal
damage and have led many to believe that methamphetamine use
may impart neurotoxic effects and as a result impair cognitive
functioning. However, given that these neuroimaging tools do not
determine causality, it is unclear whether these neural markers are
premorbid impairments, are the result of neurotoxicity, or are an
adaptive process whereby neurons develop tolerance to metham-
phetamine’s pharmacologic effects and thereby undergo physical
and structural changes to adapt to drug exposure (O’Neil et al.,
2007).
Neuropsychological Functioning in Abstinent
Methamphetamine Users
Studies that have assessed chronic methamphetamine users dur-
ing short-term abstinence (range: 4 –15 days) have yielded mixed
results for cognitive recovery. In some studies, abstinent metham-
phetamine users performed significantly worse on measures of
processing speed (Boileau et al., 2008;Kalechstein, Newton, &
Green, 2003;Simon, Dean, Cordova, Monterosso, & London,
2010), learning and memory (Boileau et al., 2008;Kalechstein et
al., 2003), attention and psychomotor speed (Kalechstein et al.,
2003), and executive functioning (Boileau et al., 2008;Kalechstein
et al., 2003). In contrast, in other studies, no significant group
differences were found on learning and working memory (Simon
et al., 2010), executive functioning (Kalechstein et al., 2003;
Simon et al., 2010), or visual spatial functioning (Kalechstein et
al., 2003).
Studies that have assessed neuropsychological performance at
longer intervals of abstinence (range: 2–36 months) have also
revealed contradictory findings. Users who maintained abstinence
exhibited higher rates of neuropsychological impairment com-
pared with controls on measures of attention, learning, and work-
ing memory (Cherner et al., 2010a;Simon et al., 2010), verbal
fluency (Chang et al., 2002), and information processing and
motor ability speed (Chen et al., 2015). In contrast, groups did not
differ on measures of motor ability (Johanson et al., 2006), verbal
fluency (Cherner et al., 2010a), attention, and verbal memory
(Rippeth et al., 2004).
In summary, the literature examining cognitive function in
methamphetamine users following abstinence contains discordant
findings. For example, different studies suggest variously that (a)
some cognitive functions improve following several weeks of
abstinence, whereas others present with more prolonged impair-
ment (Chang et al., 2002;Cherner et al., 2010a); (b) there is no
significant difference between methamphetamine and healthy
comparison groups, suggesting that cognitive impairment subsides
within weeks or months of abstinence (Chang et al., 2005;Ka-
lechstein et al., 2003;Simon et al., 2010); and (c) there is persistent
impairment, suggesting that cognitive dysfunction may linger for
many months following the cessation of use (Chang et al., 2002).
Therefore, it is not clear whether methamphetamine users improve
cognitively with abstinence, and if so, when and to what extent this
improvement occurs.
Limitations in the Current Body of Evidence
One factor that may account for variability in the cognitive
performance of methamphetamine users is the duration of absti-
nence. Studies of neuropsychological functioning do not typically
differentiate between short-term and long-term abstinence (i.e.,
when the clinical withdrawal symptoms of methamphetamine have
subsided). Given the difficulty of recruiting eligible participants,
most studies include participants with a wide range of abstinence
durations, ranging from a few days to many months. Studies
assessing participants during short-term abstinence present with
interpretative problems, as differences between groups may be attrib-
utable to acute withdrawal symptoms (Pope, Gruber, Hudson,
Huestis, & Yurgelun-Todd, 2001). Methamphetamine has a plasma
half-life of 12 hr, and the acute stimulant effects typically last
between 4 and 24 hr (Cook et al., 1993). However, there is no
consensus regarding how to operationalize the length of metham-
phetamine withdrawal (Pennay & Lee, 2011). Some studies define
the withdrawal phase as a few days, whereas others characterize it
as lasting from a few weeks to several months (Newton et al.,
2004). Withdrawal from methamphetamine typically produces a
constellation of symptoms, including fatigue, irritability, aggres-
sion, and paranoia (Meredith et al., 2005;Pennay & Lee, 2011),
that individually and collectively have the potential to impair
cognition. Nevertheless, withdrawal from methamphetamine (and
other stimulants) is not associated with potentially life-threatening
problems, such as seizures and altered consciousness, that accom-
pany alcohol and sedative-hypnotic drug withdrawal (Gortney et
al., 2016;Perry, 2014;Santos & Olmedo, 2017).
Most studies are limited by small sample sizes, attributable in
part to high rates of drop outs (Cook, Quinn, Heinzerling, &
Shoptaw, 2017). Additionally, some studies have not matched
groups on relevant confounding variables, including age and edu-
cation (Finlayson, Johnson, & Reitan, 1977). Some studies have
also not documented comorbid neuropsychiatric disorders, sub-
stance use disorders, or medical conditions that may affect cogni-
tive functioning. For example, methamphetamine users are at
increased risk of HIV and hepatitis C virus as a result of needle
sharing from injection drug use (Scott et al., 2007). HIV, which is
prevalent among methamphetamine users, exerts a deleterious
influence on cognitive functioning (Rippeth et al., 2004).
The Goal of the Present Review
The aim of this meta-analysis was to clarify the relationship
between abstinence and neuropsychological functioning in people
with methamphetamine use disorder (i.e., abuse or dependence).
1
In addition, the following research questions were examined to
ascertain the relationship between study-level covariates and effect
size: (a) To what extent do demographic characteristics predict
effect size differences between groups? (b) To what extent do
1
The majority of studies in the meta-analysis used DSM–IV criteria of
methamphetamine dependence, whereas four studies classified participants
with DSM–IV criteria of methamphetamine abuse or dependence. How-
ever, the DSM–5 diagnosis of Substance Use Disorder no longer distin-
guishes abuse from dependence. We use the term Methamphetamine Use
Disorder as the closest approximation to the DSM–5 Substance Use Dis-
order classification that now encompasses DSM–IV criteria of metham-
phetamine abuse or dependence.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
741
METHAMPHETAMINE META-ANALYSIS

clinical characteristics (e.g., duration of methamphetamine use and
length of abstinence) predict effect size differences between
groups? (c) To what extent do methodological moderators predict
effect size differences between groups? Demographic and clinical
predictors were guided by theory and prior literature. To date,
there is no consensus regarding the influence of the duration of
methamphetamine use and abstinence on cognitive recovery po-
tential. We predicted that longer duration of methamphetamine use
would be associated with worse cognitive performance, whereas a
longer duration of abstinence would be associated with better
cognitive performance. The potential methodological moderators
were exploratory and therefore hypothesis-generating.
There have been two published meta-analyses comparing indi-
viduals with a history of methamphetamine abuse or dependence
with a group of healthy controls (Potvin et al., 2018;Scott et al.,
2007). These reviews showed that methamphetamine use was
associated with medium-to-large cognitive deficits for most cog-
nitive domains but were most severe for learning, executive func-
tioning, memory, and processing speed. However, previous work
has neglected to incorporate a set of inclusion/exclusion criteria
that elected only studies that reported a period of abstinence. This
review goes beyond the two meta-analyses by including only
studies that reported a period of abstinence, thereby examining a
larger set of studies addressing the relationship between abstinence
and neuropsychological functioning. In addition, this is the first
meta-analysis on psychoactive substance use and neuropsycholog-
ical functioning that used an evidence-based method for grouping
neuropsychological tests into broad cognitive abilities.
Method
Protocol and Registration
The systematic review protocol was registered with the Interna-
tional Prospective Register of Systematic Reviews (PROSPERO:
CRD42018083598). The review was conducted according to
PRISMA guidelines (Moher et al., 2009).
Eligibility Criteria
Studies were eligible if (a) they included a methamphetamine
group that identified methamphetamine as their primary substance
of use, and a had history of methamphetamine abuse or depen-
dence (studies that included amphetamine participants were eligi-
ble if they reported separate statistics for the methamphetamine
group), (b) they reported period of abstinence from methamphet-
amine, (c) they included a healthy comparison participants (i.e.,
nonusers, did not have history of stimulant abuse or dependence),
(d) outcome measures included valid and reliable cognitive/neu-
ropsychological tests, and (e) they were published in English.
Studies were excluded if they were (a) animal studies, (b) case
reports, or (c) commentaries.
Information Sources and Search Strategies
Medline Ovid, PsycINFO, and EMBASE databases were
used to identify eligible studies. The following search terms
were entered into the advanced search on the 18th September
2017 to produce a search with all related items: methamphet-
amine or amphetamine; neurocog
!
or cognit
!
or neuropsychol
!
or assess
!
or abilit
!
or effect
!
or process
!
or impair
!
or func-
tion
!
; residual or abstinen
!
or abstain
!
or persist
!
or lasting. A
final full list of full-text articles was completed and updated on
December 16, 2017.
Data Screening and Extraction
To determine the degree of interrater reliability the two authors
(C.B., R.H.) independently screened 339 titles and abstracts and
showed a high level of agreement on articles to be included
(Cohen’s K !0.83; McHugh, 2012).
Titles and abstracts were screened independently by two review
authors (C.B., R.H.), who were blinded to authors and journal
names, to identify studies that met the inclusion criteria outlined
earlier, and duplicates removed. The full-text of the eligible studies
was retrieved and independently assessed by two reviewers (C.B.,
R.H.) against the inclusion criteria. Discrepancies were resolved
by a third rater. A total of 31 articles met the inclusion criteria (see
online supplemental materials file 1). The first author (C.B.) did
not utilize the reference list of included studies to identify addi-
tional studies as this procedure could have introduced citation bias
(Higgins, Green, & Cochrane Collaboration, 2008). The first au-
thor (C.B.) extracted all the data and, where required, contacted the
first author to request missing information. The authors of the
following studies did not respond to contact or could not be
contacted: Iudicello et al. (2010) and Kim et al. (2009).
For each study, the following data were extracted: (a) aim, (b)
study design, (c) funding, (d) sample size, (e) age, (f) gender, (g)
education, (h) urinalysis, (i) participant recruitment, (j) age of first
use, (k) lifetime dose of methamphetamine, (l) amount of use (g
per day; (m) duration of methamphetamine use, (n) abstinence
period, (o) history of substance use or dependence, (p) medical
history, (q) psychiatric history, and (r) summary statistics for the
calculation of effect size.
Data Items and Summary Measures
To provide an empirically validated comparison of results across
studies, neuropsychological tests were assigned to cognitive do-
mains based on the Cattell-Horn-Carroll (CHC) model. The CHC
model is the most comprehensive empirically validated description
of cognitive abilities (Jewsbury, Bowden, & Strauss, 2016;Lough-
man, Bowden, & D’Souza, 2014;McGrew, 2009). Confirmatory-
factor analytic studies have shown that the CHC framework pro-
vides a good fit for a wide variety of cognitive tests, such as the
Wechsler Intelligence Scales, Wechsler Memory Scale, Controlled
Oral Word Association, Rey Auditory Verbal Learning Test, and
Wisconsin Card Sorting Test (Jewsbury, Bowden, & Duff, 2017;
McGrew, 2009). Eleven of the 18 broad abilities in the CHC
framework were represented by included studies: fluid reasoning
(Gf); short-term working memory (Gwm); learning efficiency
(Gl); visual-spatial processing (Gv); comprehension knowledge
(Gc); retrieval fluency (Gr); processing speed (Gs); reaction and
decision speed (Gt); and psychomotor speed (Gps).
Tests were assigned to CHC broad factors based on prior liter-
ature of the factor structure of tests and on the agreement of two
independent raters (C.B. and S.B.; see online supplemental mate-
rial files 2 and 3). For most test scores, higher scores represented
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
742 BASTERFIELD, HESTER, AND BOWDEN

better performance. However, for those tests where higher scores
represented worse performance (e.g., tests assessing errors or
speed) group mean scores were multiplied by "1(Higgins et al.,
2008). When multiple CHC factors represented a test, a single
factor with the greatest loading was selected (Jewsbury et al.,
2017). In most cases authors reported data on more than one
outcome, or more than one time point, where the different out-
comes (or time points) were based on the same participants. The
inclusion of multiple effect size estimates from the same study
violates the assumption of independent data points, as dependence
between data points can pose a serious threat to the validity of
meta-analytic findings (Scammacca, Roberts, & Stuebing, 2014;
Wolf, 1986). To preserve statistical independence, effect sizes
within studies were combined based on CHC factors to provide a
single estimate of the overall effect to include in the meta-analysis
(Cooper, 1998;Scammacca et al., 2014). Therefore, when a study
employed several tests of the same factor or more than one subtest
from a single test (e.g., Stroop test, Trails), an average score was
derived. This meant that each study generated an average score for
each CHC factor that was tested. This approach was selected as it
preserves as much data from each study while minimizing any
violations of the assumption that data points are independent
(Cooper, 1998).
Risk of Bias
The QUADAS-2 tool is recommended for use in systematic
reviews to evaluate risk of bias in diagnostic accuracy studies
(Robinson & Goodman, 2013). QUADAS-2 consists of four do-
mains; namely, patient selection, index test, reference standard,
and flow and timing. Signaling questions within each domain were
tailored with consideration of the major types of systematic error
in case-control studies: (a) selection bias, (b) information bias, and
(c) confounding bias that may affect the cumulative evidence
(Kopec & Esdaile, 1990). Studies were coded by each QUADAS
outcome as “low risk of bias,” “high risk of bias,” or “unclear risk
of bias.” Two review authors (C.B., R.H.) independently assessed
the risk of study-level bias of included studies and observed a high
level of agreement (Cohen’s K!0.87; McHugh, 2012). Blind
assessments for risk of bias evaluation were not conducted because
contacting study authors to request more information is likely
subject to recall bias and may lead to overly positive answers
(Haahr & Hróbjartsson, 2006). Therefore, missing information and
incomplete reporting of methods was coded as “unclear risk of
bias.” Evaluation of study-level bias was incorporated into the
quantitative synthesis of results in terms of preplanned sensitivity
analyses to test the influence of removing poor-quality studies.
Calculation of Effect Size and Data Analysis
The meta-analysis was conducted using Comprehensive Meta-
Analysis Version 2.0 (Borenstein, Hedges, Higgins, & Rothstein,
2009a). Given the degree of variability within and between studies
(e.g., population characteristics, abstinence length), a random ef-
fects model was used (Borenstein, Hedges, Higgins, & Rothstein,
2009b). To estimate the mean distribution in a random-effects
model, study weights were assigned to minimize within and be-
tween sources of variance.
The Q-statistic was employed as measures of heterogeneity over
the popular I
2
statistic because direct interpretation of the I
2
is
ambiguous. According to Borenstein, Higgins, Hedges, and Roth-
stein (2017) the I
2
statistic is a proportion that represents the ratio
of true variance in effect size over the observed variance in effect
size. Therefore, a high I
2
could be attributable to high true variance
in effect size, the result of low observed variance, or a combination
of both. We used prediction intervals to quantify the extent of
observed heterogeneity.
The primary outcome was standardized mean difference (SMD,
calculated as Hedges’ g;Hedges & Olkin, 1985). Effect size
estimates were calculated for each of the nine cognitive factors.
The direction of effect size was set in such a way that negative
effect sizes represented worse cognitive performance in the meth-
amphetamine group. Effect size estimates of 0.2, 0.5, and 0.8 were
categorized as small, medium, and large, respectively (Cohen,
1988). Variability in effect size was examined using metaregres-
sion (for continuous predictors) and subgroup analyses (Qbet, or
between subgroup homogeneity for categorical predictors) to test
whether there was a significant relationship between each potential
moderator and effect size. The following predictors were grouped
as either (a) demographic predictors included methamphetamine age,
healthy control age, educational discrepancy (i.e., difference in years
of education between methamphetamine and healthy control groups),
(b) clinical predictors included length of abstinence and duration of
methamphetamine use, and (c) methodological predictors were
dummy coded to produce two variables. In the dummy codes, low and
unclear versus high risk of bias, respectively, were coded as 0 and 1.
Metaregression was conducted on CHC factors where more than
10 data points were available, and because of the small number of
studies, each moderator was examined separately for each factor
(Borenstein et al., 2009b). It was not possible to include gender,
age of first use, lifetime dose of methamphetamine, amount of use,
history of substance use, or medical history because too few
studies reported relevant data. The Holm-Bonferroni sequential
procedure was used to adjust the nominal significance level (i.e.,
p#.05; Gaetano, 2013;Holm, 1979) to control for Type I error
arising from multiple comparisons without inflating the risk of
Type II error as much as the standard, highly conservative Bon-
ferroni method (Schmidt, 1992). This procedure was applied
within each CHC factor for each initially significant (uncorrected
for family wise error) result.
Sensitivity analyses were conducted for cognitive factors to
determine the robustness of the current findings. This was done to
test the resiliency of pooled cognitive factors against individual
studies to control for biases in each single study. Primary analyses
were restricted to studies at low and unclear risk of bias by
removing high risk of bias studies. This threshold was selected to
balance statistical power against quality of current studies. This
threshold of study selection was selected as it was not possible to
remove studies labeled as unclear risk of bias as this would have
removed all studies from the analysis.
The majority of studies comprised participants who met Diag-
nostic and Statistical Manual of Mental Disorders, fourth edition
(DSM–IV) criteria for methamphetamine dependence. However,
four studies included participants who met DSM–IV criteria for
methamphetamine abuse or dependence. Additional sensitivity
analyses were performed by removing the four studies that com-
bined participants who met DSM–IV criteria for methamphetamine
abuse or dependence. This analysis was conducted because these
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
743
METHAMPHETAMINE META-ANALYSIS

latter studies may provide a more heterogenous sample than meth-
amphetamine dependence alone.
One potential source of Type I error facing meta-analytic studies
is publication bias, as this can result in an overestimation of effect
size in studies. Publication bias was assessed informally by visu-
ally inspecting forest plots and funnel plots, and then formally by
Egger’s regression test in which the standard normal deviate is
regressed on precision (i.e., inverse of the standard error; Boren-
stein et al., 2009b). Finally, to adjust for publication bias, the
trim-and fill-method was used to impute hypothetical missing
studies into the analysis to recalculate the effect size.
Results
Study Selection and Characteristics
From the 31 studies, there were 1,008 abstinent methamphet-
amine individuals and 984 healthy controls. The mean age was a
weighted average of 33.41 years for the methamphetamine group
and 34.76 years for the control group (d!0.2). In terms of
education, the methamphetamine group had a weighted average of
12.31 years of education and the control group 13.88 years (d!
1.31; see Table 1).
Meta-Analyses
The mean weighted effect size for each domain across 31
studies ranged from ".47 to .00 (Table 2 column 4, Table 3, and
online supplemental material file 4) standard deviation units, with
reaction time (RT) and decision speed (95% CI ["0.04, 0.41])
showing a small effect, and learning efficiency (95% CI
["0.58, "0.37]) the largest effect. The median for the weighted
effect size was "0.32. There was considerable heterogeneity be-
tween studies for all cognitive factors except learning efficiency
(PI: "0.60 to "0.34; see Table 2 column 8).
The mean unweighted effect size ranged from "0.56 to 0.00
(Table 2 column 7) standard deviation units, with fluid reasoning
showing a small effect and learning efficiency the largest effect.
The median for the unweighted effect size was "0.23.
Metaregression Analyses
For fluid reasoning, metaregression analyses revealed no signif-
icant association between age of methamphetamine participants
and effect size ($!"0.04, 95% CI ["0.17, 0.09], p!.56), nor
was age of healthy control participants significant ($!0.03, 95%
CI ["0.14, 0.20], p!.72). Similarly, length of abstinence was not
significant ($!0.00, 95% CI ["0.00 to 0.00], p!.22). Because
48% of the studies did not match methamphetamine and healthy
control groups on education, a variable representing the difference
in education between the groups was created for each study. This
analysis revealed that there was no significant relationship be-
tween educational discrepancy and effect size ($!0.14, 95% CI
["0.12, 0.39], p!.29). However, results revealed duration of
methamphetamine use ($!"0.16, 95% CI ["0.26, "0.07], p#
.001) was significantly related to the magnitude of effect size. The
beta coefficient remained statistically significant following the
Holm-Bonferroni adjustment (p!.002).
For short-term working memory there was no significant associa-
tion between age of methamphetamine participants and effect size
($!0.04, 95% CI ["0.04, 0.12], p!.36), and age of healthy control
participants ($!0.02, 95% CI ["0.07, 0.12], p!.62). Similarly,
there was no significant effect for length of abstinence ($!"0.00,
95% CI ["0.01, 0.00], p!.60), and educational discrepancy ($!
0.07, p!.29, 95% CI ["0.06, 0.19], p!.29) on the magnitude of
effect size. There were not sufficient data points to conduct a regres-
sion analysis for duration of methamphetamine use.
For learning efficiency, there was no significant effect of age of
methamphetamine participants ($!0.01, 95% CI ["0.03, 0.05],
p!.66), age of healthy control participants ($!0.02, 95% CI
["0.01, 0.07], p!.25). Similarly, length of abstinence ($!0.00,
95% CI ["0.001, 0.0025], p!.46), educational discrepancy
($!"0.04, 95% CI ["0.10, 0.02], p!.15), and duration of use
($!0.00, 95% CI ["0.0029, 0.0042], p!.73) had no significant
effect.
For comprehension knowledge, there was no significant effect
of age of methamphetamine participants ($!"0.08, 95% CI
["0.16, 0.0049], p!.07), and age of healthy control participants
($!"0.03, 95% CI ["0.11, 0.06], p!.53). However, length of
abstinence had a significant effect ($!"0.00, 95% CI
[0.0002, "0.0008], p#001). The beta coefficient remained sta-
tistically significant after the Holm-Bonferroni adjustment (p#
.001). For educational discrepancy ($!"0.30, 95% CI ["0.65,
0.05], p!.09) and duration of use ($!"0.00, 95% CI ["0.0078,
0.0005], p!.09) there was no significant effect.
For processing speed, there was no significant effect of age of
methamphetamine participants ($!0.06, 95% CI ["0.02, 0.16],
p!.12), age of healthy control participants ($!0.00, 95% CI
["0.09, 0.09], p!.99), or length of abstinence ($!0.00, 95% CI
["0.00, 0.00], p!.18). Similarly, for educational discrepancy
($!0.02, 95% CI ["0.16, 0.20], p!.83) and duration of
methamphetamine use ($!"0.01, 95% CI ["0.05, 0.027], p!
.52) there was no significant effect (see Table 4).
2
2
In a secondary and more conservative set of analyses, we conducted the
Holm-Bonferroni correction across rather than within CHC factors. The
overall pattern of results remained unchanged. For fluid reasoning, the beta
coefficient again remained significant (p!0.003) for duration of meth-
amphetamine use. For comprehension knowledge, the beta coefficient
again remained significant following the Holm-Bonferroni adjustment
(p#.001) for length of abstinence.
Table 1
Participants’ Demographic Data and Methamphetamine
Use Characteristics
Characteristic k
MA group HC group
nM(SD)nM(SD)
Age 31 1203 33.41 (6.78) 1102 34.76 (7.25)
% Male 23 985 69 1027 66
% Female 23 985 32 1027 34
Education (years) 28 1046 12.31 (.91) 1067 13.88 (1.43)
Length of MA use (months) 27 984 144 (5.3)
Length of MA abstinence
(days) 31 1070 128.8 (99.5)
Note.k!number of studies; MA !methamphetamine; n!number of
participants; HC !healthy control.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
744 BASTERFIELD, HESTER, AND BOWDEN

Subgroup Analyses
There was no difference between the groups in terms of their
risk of bias classification for fluid reasoning (Qbet !0.15, df !1,
p!.70), short-term working memory (Qbet !0.004, df !1, p!
.95), learning efficiency (Qbet !0.53, df !1, p!.47), visual-
spatial processing (Qbet !1.64, df !1, p!.20), retrieval fluency
(Qbet !0.76, df !1, p!.39), processing speed (Qbet !0.81,
df !1, p!.37), or psychomotor speed (Qbet !0.073 df !1, p!
.39). However, for comprehension knowledge, there was a statis-
tically significant difference between the groups (i.e., unclear vs.
high risk of bias: Qbet !7.69, df !1, p!.006). As such, the high
risk of bias studies demonstrated a negative medium effect in favor
of the control group ("0.67 SD units; 95% CI ["0.92 to "0.43),
compared with the unclear risk of bias studies where the confi-
dence interval for the average effect size overlaps with zero
("0.09 SD units; 95% CI ["0.42, 0.24]). There were not sufficient
data points to conduct a meaningful subgroup analysis for reaction
and decision speed (see Table 5).
3
Publication Bias
Visual inspection of funnel plots and Egger’s test revealed no
evidence of publication bias for fluid reasoning (t!0.00350, df !
12, p!.99), visual-spatial processing (t!1.63, df !3.00, p!
.10), comprehension knowledge (t!1.72, df !10.00, p!.06),
retrieval fluency (t!1.66, df !7.00, p!.14), processing speed
(t!0.07, df !20.00, p!.95), reaction and decision speed (t!
1.01, df !1.00, p!.50), or psychomotor speed (t!0.62, df !
7.00, p!.56).
Short-term working memory (t!2.19, df !10.00, p!.05) and
learning efficiency (t!2.76, df !15.00, p!.01) were potentially
influenced by publication bias. Using the trim and fill method, the
effect size for short-term working memory remained nonsignifi-
cant but was reduced in magnitude ("0.04 SD units, 95% CI
["0.26, 0.18]) and for learning efficiency the effect size remained
significant and unchanged in magnitude ("0.40 SD units, 95% CI
["0.52, "0.28]; see online supplemental material file 5).
Sensitivity Analyses
The removal of high risk of bias studies did not change the
results, underscoring the robustness of findings (see Table 6 and
Figure 1). However, the result of a smaller set of studies per
analysis meant that there was greater sampling variation and hence
the reduced level of precision was reflected in the larger confi-
dence intervals. For comprehension knowledge, fluid reasoning,
and visual-spatial processing the confidence intervals for the av-
erage effect sizes overlapped with zero.
The removal of four studies that met DSM–IV criteria for abuse
or dependence did not change the results. For fluid reasoning,
short-term working memory, learning efficiency, retrieval fluency,
processing speed, and psychomotor speed, the confidence intervals
for the average effect sizes overlapped with zero (see online
supplemental material file 6).
3
The Holm-Bonferroni sequential correction was applied across CHC
factors for the subgroup analyses. For comprehension knowledge, the beta
coefficient remained significant following the Holm-Bonferroni adjust-
ment (p!.05).
Table 2
Summary of Meta-Analytic Effect Sizes and Sensitivity Analyses for CHC Domain
Cognitive domain k
1
MA NHC N
Hedges’ g
(weighted)
SD
(weighted)
95% CI
(weighted)
PI
(weighted)
Hedges’ g
(unweighted) QI
2
%
2
t
Egger’s
test (t)
Fluid reasoning 14 423 475 ".02 .22 [".46, .42] ["2.04, 1.45] .00 148.105 91.22 .61 .78 .00
Short-term working memory 12 333 376 ".19 .13 [".44, .06] ["1.01, 1.01] ".22 32.415 66.07 .12 .35 2.19
Learning efficiency 17 657 673 ".47 .05 [".58, ".37] [".60, ".34] ".56 16.180 1.11 .00 .02 2.76
Visual-spatial processing 5 157 147 ".42 .16 [".74, ".11] ["1.39, .54] ".48 8.452 52.68 .07 .26 1.63
Comprehension knowledge 12 357 325 ".43 .13 [".69, ".18] ["1.32, .44] ".41 35.308 68.85 .14 .37 1.72
Retrieval fluency 9 284 324 ".32 .14 [".59, ".04] ["1.18, .54] ".36 21.447 62.70 .11 .33 1.66
Processing speed 22 696 692 ".28 .12 [".52, ".04] ["1.34, .78] ".11 111.582 81.18 .24 .49 .07
Reaction time and decision
speed 3 47 47 .00 .21 .00 ["2.74, 2.73] .00 .000 .00 .00 .00 1.01
Psychomotor speed 9 291 333 ".35 .15 [".65, ".05] ["1.30, .60] ".37 24.937 67.92 .14 .37 .62
Total k/N 31 6,637
Note. MA !methamphetamine; HC !healthy control. k
1
refers to the overall number of studies in the meta-analysis, pooled sample size (N), weighted
effect size, standard deviation, 95% weighted confidence interval, weighted prediction interval, 95% unweighted confidence interval, and homogeneity (Q,
I
2
,%
2
,t), publication bias, and sensitivity analyses.
Table 3
Stem and Leaf Plot of Weighted Effect Sizes
Stem Leaf
".5
".4 2, 3, 7
".3 2, 5
".2 8
".1 9
.0 0, 2
.1
.2
.3
.4
.5
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
745
METHAMPHETAMINE META-ANALYSIS

Discussion
Summary of Results
This systematic review and meta-analysis assessed the evidence
pertaining to the relationship between abstinence and methamphet-
amine use and identified potential moderators of effect size. The
review identified 31 studies examining the relationship between
abstinence and neuropsychological functioning in people with a
history of dependent methamphetamine use compared with healthy
controls.
As noted earlier, methamphetamine use acts primarily via dopami-
nergic and serotonergic frontostriatal pathways (Mark, Soghomonian,
&Yamamoto,2004), and these alterations may lead to impaired
neuropsychological functioning in abstinent methamphetamine users.
The results of the meta-analysis broadly support this hypothesis by
showing small-to-moderate group differences in several cognitive
domains dependent on frontostriatal circuits, including learning effi-
ciency, visual-spatial processing, comprehension knowledge, retrieval
fluency, processing speed, and psychomotor speed (see Table 2 and
Figures 4.1 to 4.9 in the online supplemental material). Three excep-
tions, revealing no significant group differences, were in domains of
fluid reasoning, short-term working memory, and reaction and deci-
sion speed.
The significance of between-study variance prompted an investi-
gation of whether age, education level, length of abstinence, and
duration of methamphetamine use could account statistically for some
of the between-study dispersion. This variation was expected to be
associated with differences in length of abstinence and duration of
methamphetamine use. Therefore, the next step was to identify the
study characteristics most strongly associated with effect size. This
meta-analysis revealed that duration of methamphetamine use bore a
significant relation to only the fluid reasoning effect size, which,
paradoxically, was not impaired, on average (see Table 2).
Table 4
Relationship Between Study Characteristics and Neuropsychological Outcomes
Study characteristic kQ model pvalue of Q model Q residual pvalue of Q residual $
Fluid reasoning
Demographic predictors
MA age 14 .34 .56 145.19 .001 ".04
HC age 14 .13 .72 145.47 .001 .03
Educational discrepancy 11 1.14 .29 122.43 .001 .14
Clinical predictors
Length of abstinence 11 1.49 .22 129.25 .001 .00
Duration of MA use 11 12.47 #.0001
!!!
60.77 .00 ".16
Short-term working memory
Demographic predictors
MA age 12 1.04 .36 31.61 .001 .04
HC age 12 .24 .62 30.38 .001 .02
Educational discrepancy 11 1.11 .29 23.09 .006 .07
Clinical predictors
Length of abstinence 11 .28 .60 24.45 .001 ".00
Duration of MA use 8 — — — — —
Comprehension knowledge
Demographic predictors
MA age 11 3.39 .07 22.22 .008 ".08
HC age 11 .40 .53 30.03 .001 ".03
Educational discrepancy 11 2.80 .09 27.06 .001 ".30
Clinical predictors
Length of abstinence 11 27.58 #.001
!!!
7.70 .57 ".00
Duration of MA use 11 2.90 .09 21.85 .005 ".00
Processing speed
Demographic predictors
MA age 22 2.43 .12 108.15 .001 .06
HC age 22 .00 .99 110.31 .001 .00
Educational discrepancy 21 .05 .83 111.51 .001 .02
Clinical predictors
Length of abstinence 22 1.83 .18 105.58 .001 .00
Duration of MA use 18 .41 .52 105.88 .001 ".01
Learning efficiency
Demographic predictors
MA age 16 .19 .66 15.90 .32 .01
HC age 16 1.30 .25 14.75 .40 .02
Educational discrepancy 15 2.12 .15 13.09 .44 ".04
Clinical predictors
Length of abstinence 15 .55 .46 14.69 .33 .00
Duration of MA use 12 .12 .73 8.76 .55 .00
Note. Weighted random effects analyses; MA !methamphetamine; HC !healthy control; k!number of studies; $!dregression weight (beta).
!
p#.05.
!!
p#.01.
!!!
p#.001.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
746 BASTERFIELD, HESTER, AND BOWDEN

There is no consensus regarding the influence of duration of
methamphetamine use on neuropsychological functioning (Hoff-
man et al., 2006;Johanson et al., 2006). The inconsistent findings
may, in part, be attributable to the reliance on self-report measures
of methamphetamine use. In addition, methamphetamine purity
varies within and between geographical populations (Community
Epidemiology Work Group, National Institute of Drug Abuse,
2006), so it is unclear whether duration of use is consistent across
populations. Nevertheless, the current study found that magnitude
of the group difference increased with every point (year) increase
in duration of methamphetamine use by 0.16 SD units, in favor of
the healthy control group for fluid reasoning (see Table 4). The
data support the hypothesis that longer duration of methamphet-
amine use is associated with greater decrements in cognitive
functions (Sekine et al., 2003).
Potential Biases in the Review Process
Methodological problems within primary studies included in
this meta-analysis may limit the interpretation of our results. For
example, there was a high risk of bias in some studies included in
the analysis (see Figure 1 and online supplemental material file 6).
The participant inclusion processes in some studies may have
given rise to selection bias, constraining the matching of cases and
controls. Selection bias is a particular problem in case-control
studies, as it gives rise to noncomparability between cases and
controls (Hennekens & Buring, 1987). In addition, a high risk of
bias across studies was evident in the rigor of the reference
standard (e.g., urinalysis). For example, 15 studies did not report
whether the healthy controls were subject to urinalysis, increasing
the potential for measurement bias. Measurement bias results from
systematic differences in the way data are collected from cases and
controls, and arises when study variables are inaccurately mea-
sured or classified (Greenhalgh, 2001). The comprehension-
knowledge domain showed a significant effect of bias ratings on
group differences (i.e., unclear vs. high risk of bias). The studies at
high risk of bias demonstrated a medium effect in favor of the
control group, compared with those classified as unclear risk, for
which the confidence interval for the average effect size over-
Table 6
Sensitivity Analyses
Cognitive domains k
1
Hedges’ g
Sensitivity analyses
(high risk of bias removed) k
2
Hedges’ g
Sensitivity analyses
(MA abuse or dependence removed)
Fluid reasoning 6 ".172 ["1.049, .704] 11 ".220 [".713, .273]
Short-term working memory 5 ".283 [".524, ".042] 10 ".159 [".442, .124]
Learning efficiency 7 ".457 [".699, ".246] 14 ".452 [".565, ".340]
Visual-spatial processing 3 ".250 [".555, .054] 5 — —
Comprehension knowledge 5 ".082 [".326, .162] 12 — —
Retrieval fluency 6 ".338 [".661, ".014] 8 ".213 [".457, .031]
Processing speed 9 ".573 [".992, ".153] 20 .126 [".467, .027]
Reaction time and decision speed 1 — — 3 — —
Psychomotor speed 5 ".502 [".818, ".186] 7 ".311 [".677, .054]
Note.k
1
refers to the number of studies in the sensitivity analysis once the high risk of bias studies were removed. k
2
refers to the number of studies in
the sensitivity analysis once the studies that included participants with a diagnosis of MA abuse or dependence were removed, pooled effect size (Hedges’
g), 95% CI by cognitive domain.
Table 5
Subgroup Analyses of Cognitive Domains Controlling for Risk of Bias Classification
Cognitive domain Moderator kHedges’ g95%CI
Fluid reasoning Unclear risk 6 ".12 [".81, .56]
High risk 8 .06 [".53, .64]
Short-term working memory Unclear risk 4 ".18 [".63, .27]
High risk 8 ".20 [".51, .12]
Learning efficiency Unclear risk 7 ".43 [".59, ".26]
High risk 10 ".51 [".64, ".37]
Visual-spatial processing Unclear risk 3 ".27 [".64, .11]
High risk 2 ".67 ["1.16, ".18]
Comprehension knowledge Unclear risk 5 ".09 [".42, .24]
High risk 7 ".67 [".92, ".43]
Retrieval fluency Unclear risk 4 ".18 [".60, .24]
High risk 5 ".44 [".82, ".05]
Processing speed Unclear risk 8 ".42 [".82, ".03]
High risk 14 ".19 [".50, .11]
Psychomotor speed Unclear risk 5 ".47 [".88, .07]
High risk 4 ".21 [".65, .23]
Note. A random effects analysis with a pooled variance estimate was used to calculate subgroup analyses.
Because we had a small number of studies per subgroup a pooled variance estimate meant that we could obtain
a more accurate estimate of tau
2
than one based on a smaller subset of studies, which would have introduced
more error.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
747
METHAMPHETAMINE META-ANALYSIS

lapped with zero. Therefore, the proportion of information from
studies at high risk of bias may be sufficient to distort the inter-
pretation of the results. Nevertheless, sensitivity analyses revealed
that for the majority of cognitive domains, the results remained
unchanged when high risk of bias studies were removed, suggest-
ing that the findings may be sufficiently robust (see Table 6).
The analysis of the 31 studies (see online supplemental material
file 6) indicated an additional source of researcher bias; namely,
the extent authors were blinded to case concealment during as-
sessment of the index test (i.e., cognitive tests). For example,
researchers may have expected the control group to perform better
than the methamphetamine group, thereby affecting their perfor-
mance in subtle ways (Higgins et al., 2008).
The current meta-analysis highlighted potential publication bias
in two cognitive domains, suggesting that the performance of
methamphetamine participants may have been slightly overesti-
mated. However, with the trim and fill method, the results for
short-term working memory did not change appreciably. Never-
theless, detecting publication bias through funnel plots is some-
what subjective, and publication bias is not the only explanation
for funnel plot asymmetry (Sterne, Gavaghan, & Egger, 2000).
Alternative explanations for this asymmetry include the possibility
that smaller studies differ systematically in some way (e.g., the
level of substance pathology among participants) from larger stud-
ies (Sterne et al., 2000). Moreover, because Egger’s regression test
is based on the funnel plot, whereby the standardized effect esti-
mate is regressed on the measure of precision (standard error), this
test is subject to error when there is considerable between-study
heterogeneity (Borenstein et al., 2009b).
Summary and Conclusions
The results need to be considered in the light of several impor-
tant caveats. A major limitation was the availability for inclusion
of only cross-sectional designs. Although the results were consis-
tent with previous findings of an association between abstinence
and neuropsychological functioning (Chang et al., 2002;Henry et
al., 2011;Johanson et al., 2006), the methods cannot address the
critical question of the direction of causality. For example, we
cannot distinguish whether methamphetamine use leads to long-
term neuropsychological impairment via structural or functional brain
changes, or whether preexisting deficits in neuropsychological per-
formance and cortical integrity are vulnerability factors for initiating
methamphetamine use or transitioning to methamphetamine depen-
dence.
Previous research has found an association between duration of
methamphetamine use and both the severity of structural brain
abnormalities in frontostriatal circuits and the effect size of neu-
ropsychological deficits (Volkow et al., 2001). Similarly, there is
growing evidence in the drug dependence and gambling addiction
literatures that premorbid cortical and neuropsychological differ-
ences may be vulnerabilities to abuse and dependence (Bjork &
Grant, 2009;Schumann et al., 2010;Tarter et al., 2003;Tarter,
Kirisci, Habeych, Reynolds, & Vanyukov, 2004). For example,
deficits in inhibitory control and emotional regulation (Jordan &
Andersen, 2017), as well as certain personality traits (e.g., impul-
sivity, sensation-seeking; Jordan & Andersen, 2017), can both (a)
precede and perhaps contribute to methamphetamine use and (b)
affect neuropsychological performance. In this light, cross-sectional
designs may lack the methodological rigor to detect neuropsycholog-
ical improvement in relation to abstinence. Large-scale longitudinal
studies will be needed to examine people before they enter a period of
risk of methamphetamine use so that they can control for the con-
founding influence of premorbid cortical and neuropsychological
differences on test scores.
A limitation of most studies included in this meta-analysis was
the lack of premorbid ability measures prior to participants’ onset
of methamphetamine use. Without premorbid measures, it is dif-
ficult to rule out the possibility that the lack of significant differ-
ence between the two groups may have resulted from the enhanced
effect of methamphetamine use on attention, which in turn could
improve performance on certain neuropsychological tasks. Never-
theless, although there is evidence of acute benefits of methamphet-
amine on cognition among chronic and occasional users (Hart, Haney,
Foltin, & Fischman, 2002;Hart et al., 2008;Johnson et al., 2000;
Mahoney, Jackson, Kalechstein, De La Garza, & Newton, 2011),
there is no evidence of which we are aware that this benefit persists
beyond withdrawal. Given that the studies in the meta-analysis in-
cluded methamphetamine-using participants who were currently ab-
stinent, it is unclear what, if any, benefit might be observed from past
use to current cognition. Because we did not have access to premorbid
ability measures, the present review treated education as a partial
proxy for premorbid cognitive ability. Despite the lower educational
attainment in the methamphetamine group relative to the healthy
control group, education was not a significant predictor of effect size.
Nevertheless, this difference limits the inferences that can be drawn
between groups, as cognitive measures are influenced by educational
attainment. However, some evidence suggests that education levels
may not be an appropriate measure of premorbid IQ for dependent
drug users as other groups, because they typically initiate use while in
secondary school, in turn negatively influencing their educational
attainment (Lynskey & Hall, 2000). It is possible, however, that the
educational difference between the groups in the study (12.3 to 13.9
years) represents an especially large functional difference. The leap
from high school to college or university is a major life transition
characterized by a number of psychological changes, some of which
may decrease risk for poor performance (e.g., learning to function as
an independent adult), whereas others may increase risk for poor
performance (e.g., alcohol and drug use) on neuropsychological tasks
(Parker, Summerfeldt, Hogan, & Majeski, 2004;Perry, Hladkyj,
Pekrun, & Pelletier, 2001;White et al., 2006). Future studies should
Figure 1. Risk of bias graph. See the online article for the color version
of this figure.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
748 BASTERFIELD, HESTER, AND BOWDEN

recruit a comparison group that is closely matched to cases on poten-
tial confounding variables to control for extraneous influences on
cognition.
Cross-sectional designs are also marked by difficulty in differ-
entiating the influence of methamphetamine use from that of
confounding variables that influence cognition. Many of the 31
studies in the meta-analysis did not provide sufficient data on
potential confounding factors such as quantity of use, route of
administration, the extent of other drug use factors, personality
traits such as impulsivity, and comorbid neuropsychiatric disor-
ders. Therefore, differences between the groups may be attribut-
able to one or more third (uncontrolled) variables exerting an
influence on cognitive functioning. It will be important for future
studies to provide standardized and consistent information on these
variables so that one can more accurately assess the relationship
between methamphetamine use and brain functioning.
Taken together, this is the first meta-analysis to investigate the
relationship between abstinence and neuropsychological function-
ing in methamphetamine use using strict inclusion/exclusion cri-
teria that included only studies that reported a period of abstinence,
distinguishing this review from the previous two meta-analyses
(Potvin et al., 2018;Scott et al., 2007). Our findings differ from
those in the two previous meta-analyses examining neuropsycho-
logical deficits in methamphetamine abuse or dependence in the
magnitude of deleterious effects. The previous reviews suggested
that methamphetamine use was associated with medium-to-large
cognitive deficits for most cognitive domains but were most severe
for learning, executive functioning, memory, and processing speed.
However, these reviews included studies that did not report a
period of abstinence. This analytic decision may have resulted in
the inclusion of people who were not abstinent, experiencing the
very acute effects of withdrawal, or both, thereby potentially
inflating the effect size. Our review differs from the previous
meta-analyses by including only studies that reported a period of
abstinence, thereby examining a larger set of studies addressing the
relationship between abstinence and neuropsychological function-
ing. However, a number of study limitations, present in all three
reviews, present challenges in interpreting these findings. In par-
ticular, the participants retained in the methamphetamine group
may have been in better physical, mental, and emotional health
than those who dropped out of the study or did not participate.
Therefore, the small-to-moderate differences between the meth-
amphetamine and healthy control group may have been somewhat
attenuated as a result of participant attrition in the methamphet-
amine group. In addition, this meta-analysis suggests that strong
statements regarding cognitive functioning in abstinent metham-
phetamine users are premature given that cross-sectional studies
are unable to differentiate cognitive weaknesses that may predate
methamphetamine use from those that may be a function of meth-
amphetamine use. In general, this analysis was based on studies of
mixed methodological quality. It will be important for future
studies to improve methodological rigor of case– control studies,
especially by avoiding selection and information biases. In spite of
these limitations, the robustness of the findings is supported by the
conservative analytic approach used, for example, by undertaking
a random-effects meta-analysis that results in larger confidence
intervals and makes it more difficult to reject the null (Borenstein
et al., 2009b).
Large-scale longitudinal studies are needed to examine people
before they enter a period of risk of methamphetamine use so that
they can control for the potential confounding influence of pre-
morbid cortical and neuropsychological differences on test scores.
In addition, studies of identical twins discordant for methamphet-
amine use, although pragmatically challenging to conduct, could
help in determining the direction of causality to ascertain whether
preexisting brain abnormalities predate or follow the onset of
methamphetamine use. Finally, authors of meta-analysis com-
monly grouped cognitive tasks based on clinical judgment. This
method of test classification is invariably subject to selection and
outcome reporting bias (Pase & Stough, 2014). Future investiga-
tions in this domain would benefit from using the CHC model to
report cognitive outcomes to enable greater consistency and inter-
pretation of conclusions.
References
Albers, D. S., & Sonsalla, P. K. (1995). Methamphetamine-induced hy-
perthermia and dopaminergic neurotoxicity in mice: Pharmacological
profile of protective and nonprotective agents. The Journal of Pharma-
cology and Experimental Therapeutics, 275, 1104 –1114.
Bjork, J. M., & Grant, S. J. (2009). Does traumatic brain injury increase
risk for substance abuse? Journal of Neurotrauma, 26, 1077–1082.
http://dx.doi.org/10.1089/neu.2008.0849
Blumenfeld, H. (2010). Neuroanatomy through clinical cases. Sunderland,
MA: Sinauer.
Boileau, I., Rusjan, P., Houle, S., Wilkins, D., Tong, J., Selby, P.,...Kish,
S. J. (2008). Increased vesicular monoamine transporter binding during
early abstinence in human methamphetamine users: Is VMAT2 a stable
dopamine neuron biomarker? The Journal of Neuroscience, 28, 9850 –
9856. http://dx.doi.org/10.1523/JNEUROSCI.3008-08.2008
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R.
(2009a). Comprehensive meta-analysis (Version 2.0) [Computer soft-
ware]. Englewood, NJ: Biostat.
Borenstein, M., Hedges, L. V., Higgins, J. P. Y., & Rothstein, H. R.
(2009b). Introduction to Meta-Analysis. (1st ed.). West Sussex, UK:
Wiley.
Borenstein, M., Higgins, J. P., Hedges, L. V., & Rothstein, H. R. (2017).
Basics of meta-analysis: I2 is not an absolute measure of heterogeneity.
Research Synthesis Methods, 8, 5–18.
Bowyer, J., & Holson, R. (1995). Methamphetamine and amphetamine
neurotoxicity. In L. Chang & R. Dyers (Eds.), Handbook of neurotoxi-
cology (Vol. 2, pp. 845– 870). New York, NY: Marcel Dekker.
Chang, L., Cloak, C., Patterson, K., Grob, C., Miller, E. N., & Ernst, T.
(2005). Enlarged striatum in abstinent methamphetamine abusers: A
possible compensatory response. Biological Psychiatry, 57, 967–974.
http://dx.doi.org/10.1016/j.biopsych.2005.01.039
Chang, L., Ernst, T., Speck, O., Patel, H., DeSilva, M., Leonido-Yee, M., &
Miller, E. N. (2002). Perfusion MRI and computerized cognitive test ab-
normalities in abstinent methamphetamine users. Psychiatry Research, 114,
65–79. http://dx.doi.org/10.1016/S0925-4927%2802%2900004-5
Chen, C. K., Lin, S. K., Chen, Y. C., Huang, M. C., Chen, T. T., Ree, S. C.,
& Wang, L. J. (2015). Persistence of psychotic symptoms as an indicator
of cognitive impairment in methamphetamine users. Drug and alcohol
dependence, 148, 158 –164.
Cherner, M., Suarez, P., Casey, C., Deiss, R., Letendre, S., Marcotte, T.,
. . . the HNRC Group. (2010a). Methamphetamine use parameters do not
predict neuropsychological impairment in currently abstinent dependent
adults. Drug and Alcohol Dependence, 106 (2–3), 154 –163. http://dx
.doi.org/10.1016/j.drugalcdep.2009.08.010
Cherner, M., Suarez, P., Casey, C., Deiss, R., Letendre, S., Marcotte, T.,
. . . Heaton, R. K. (2010b). Methamphetamine use parameters do not
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
749
METHAMPHETAMINE META-ANALYSIS

predict neuropsychological impairment in currently abstinent dependent
adults. Drug and Alcohol Dependence, 106, 154 –163. http://dx.doi.org/
10.1016/j.drugalcdep.2009.08.010
Cohen, J. (1988). Statistical power analysis for the behavioural sciences.
Hillsdale, NJ: Erlbaum.
Community Epidemiology Work Group, National Institute of Drug Abuse.
(2006). Epidemiologic trends in drug abuse. (NIH Publication No.
06 –5878). Bethesda, MD: U.S. Department of Health and Human Ser-
vices. Retrieved from https://archives.drugabuse.gov/sites/default/files/
advreport106.pdf
Cook, C. E., Jeffcoat, A. R., Hill, J. M., Pugh, D. E., Patetta, P. K., Sadler,
B. M., . . . Perez-Reyes, M. (1993). Pharmacokinetics of methamphet-
amine self-administered to human subjects by smoking S-(&)-
methamphetamine hydrochloride. Drug Metabolism and Disposition,
21, 717–723.
Cook, R., Quinn, B., Heinzerling, K., & Shoptaw, S. (2017). Dropout in
clinical trials of pharmacological treatment for methamphetamine de-
pendence: The role of initial abstinence. Addiction, 112, 1077–1085.
http://dx.doi.org/10.1111/add.13765
Cooper, H. (1998). Synthesizing research: A guide for literature reviews
(3rd ed.). Thousand Oaks, CA: Sage.
Davidson, C., Gow, A. J., Lee, T. H., & Ellinwood, E. H. (2001). Meth-
amphetamine neurotoxicity: Necrotic and apoptotic mechanisms and
relevance to human abuse and treatment. Brain Research Reviews, 36,
1–22. http://dx.doi.org/10.1016/S0165-0173(01)00054-6
Dean, A. C., Groman, S. M., Morales, A. M., & London, E. D. (2013). An
evaluation of the evidence that methamphetamine abuse causes cogni-
tive decline in humans. Neuropsychopharmacology, 38, 259 –274. http://
dx.doi.org/10.1038/npp.2012.179
Drug Enforcement Administration. (2017). National drug threat assess-
ment. Retrieved from https://www.dea.gov/sites/default/files/2018-07/
DIR-040-17_2017-NDTA.pdf
Erinoff, L. (1995). General considerations in assessing neurotoxicity using
neuroanatomical methods. Neurochemistry International, 26, 111–114.
http://dx.doi.org/10.1016/0197-0186(94)00105-4
Ernst, T., Chang, L., Leonido-Yee, M., & Speck, O. (2000). Evidence for
long-term neurotoxicity associated with methamphetamine abuse: A 1H
MRS study. Neurology, 54, 1344 –1349. http://dx.doi.org/10.1212/WNL
.54.6.1344
Farhadian, M., Akbarfahimi, M., Hassani Abharian, P., Hosseini, S. G., &
Shokri, S. (2017). Assessment of executive functions in methamphetamine-
addicted individuals: Emphasis on duration of addiction and abstinence.
Basic and Clinical Neuroscience, 8, 147–154. http://dx.doi.org/10.18869/
nirp.bcn.8.2.147
Fassbender, C., Lesh, T. A., Ursu, S., & Salo, R. (2015). Reaction time
variability and related brain activity in methamphetamine psychosis. Bio-
logical Psychiatry, 77, 465– 474. http://dx.doi.org/10.1016/j.biopsych.2014
.07.028
Finlayson, M. A., Johnson, K. A., & Reitan, R. M. (1977). Relationship of
level of education to neuropsychological measures in brain-damaged and
non-brain-damaged adults. Journal of Consulting and Clinical Psychol-
ogy, 45, 536 –542. http://dx.doi.org/10.1037/0022-006X.45.4.536
Gaetano, J. (2013). Holm-Bonferroni sequential correction: An EXCEL
calculator. Retrieved from https://www.researchgate.net/publication/
236969037_Holm-Bonferroni_Sequential_Correction_An_EXCEL_
Calculator
Gortney, J. S., Raub, J. N., Patel, P., Kokoska, L., Hannawa, M., &
Argyris, A. (2016). Alcohol withdrawal syndrome in medical patients.
Cleveland Clinic Journal of Medicine, 83, 67–79. http://dx.doi.org/10
.3949/ccjm.83a.14061
Greenhalgh, T. (2001). How to read a paper: The basics of evidence based
medicine. British Medical Journal, 315, 3062–3073. http://dx.doi.org/
10.1210/en.2010-1474
Haahr, M. T., & Hróbjartsson, A. (2006). Who is blinded in randomized
clinical trials? A study of 200 trials and a survey of authors. Clinical
Trials, 3, 360 –365. http://dx.doi.org/10.1177/1740774506069153
Hart, C. L., Gunderson, E. W., Perez, A., Kirkpatrick, M. G., Thurmond, A.,
Comer, S. D., & Foltin, R. W. (2008). Acute physiological and behavioral
effects of intranasal methamphetamine in humans. Neuropsychopharmacol-
ogy, 33, 1847–1855. http://dx.doi.org/10.1038/sj.npp.1301578
Hart, C. L., Haney, M., Foltin, R. W., & Fischman, M. W. (2002). Effects
of the NMDA antagonist memantine on human methamphetamine dis-
crimination. Psychopharmacology, 164, 376 –384. http://dx.doi.org/10
.1007/s00213-002-1225-9
Hart, C. L., Marvin, C. B., Silver, R., & Smith, E. E. (2012). Is cognitive
functioning impaired in methamphetamine users? A critical review. Neuro-
psychopharmacology, 37, 586 – 608. http://dx.doi.org/10.1038/npp.2011
.276
Hart, C. L., Ward, A. S., Haney, M., Foltin, R. W., & Fischman, M. W.
(2001). Methamphetamine self-administration by humans. Psychophar-
macology, 157, 75– 81. http://dx.doi.org/10.1007/s002130100738
Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis.
New York, NY: Academic Press.
Hennekens, C., & Buring, J. (1987). Epidemiology in medicine. New York,
NY: Lippincott Williams & Wilkins.
Henry, B. L., Minassian, A., & Perry, W. (2010). Effect of methamphet-
amine dependence on everyday functional ability. Addictive Behaviors,
35, 593–598. http://dx.doi.org/10.1016/j.addbeh.2010.01.013
Henry, B. L., Minassian, A., van Rhenen, M., Young, J. W., Geyer, M. A.,
Perry, W., . . . the Translational Methamphetamine AIDS Research
Center (TMARC) Group. (2011). Effect of methamphetamine depen-
dence on inhibitory deficits in a novel human open-field paradigm.
Psychopharmacology, 215, 697–707. http://dx.doi.org/10.1007/s00213-
011-2170-2
Higgins, J. P. T., Green, S., & the Cochrane Collaboration. (2008). Co-
chrane handbook for systematic reviews of interventions. Retrieved from
http://books.google.com/books?id!s11gCx4LS_EC.http://dx.doi.org/
10.1002/9780470712184
Hoffman, W. F., Moore, M., Templin, R., McFarland, B., Hitzemann, R. J.,
& Mitchell, S. H. (2006). Neuropsychological function and delay dis-
counting in methamphetamine-dependent individuals. Psychopharma-
cology, 188, 162–170. http://dx.doi.org/10.1007/s00213-006-0494-0
Holm, S. (1979). A simple sequentially rejective multiple test procedure.
Scandinavian Journal of Statistics, 6, 65–70.
Homer, B. D., Halkitis, P. N., Moeller, R. W., & Solomon, T. M. (2013).
Methamphetamine use and HIV in relation to social cognition. Journal
of Health Psychology, 18, 900 –910. http://dx.doi.org/10.1177/1359
105312457802
Iudicello, J. E., Woods, S. P., Vigil, O., Scott, J. C., Cherner, M., Heaton,
R. K., . . . HIV Neurobehavioral Research Center (HNRC) Group.
(2010). Longer term improvement in neurocognitive functioning and
affective distress among methamphetamine users who achieve stable
abstinence. Journal of clinical and experimental neuropsychology, 32,
704 –718.
Jackson, B. J. (2008). Cognitive function and excessive daytime sleepiness
in methamphetamine-dependent individuals. Los Angeles, CA: UCLA.
Jernigan, T. L., Gamst, A. C., Archibald, S. L., Fennema-Notestine, C.,
Mindt, M. R., Marcotte, T. D.,...Grant, I. (2005). Effects of meth-
amphetamine dependence and HIV infection on cerebral morphology.
The American Journal of Psychiatry, 162, 1461–1472. http://dx.doi.org/
10.1176/appi.ajp.162.8.1461
Jewsbury, P. A., Bowden, S. C., & Duff, K. (2017). The Cattell–Horn–
Carroll model of cognition for clinical assessment. Journal of Psychoe-
ducational Assessment, 35, 547–567. http://dx.doi.org/10.1177/
0734282916651360
Jewsbury, P. A., Bowden, S. C., & Strauss, M. E. (2016). Integrating the
switching, inhibition, and updating model of executive function with the
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
750 BASTERFIELD, HESTER, AND BOWDEN

Cattell-Horn-Carroll model. Journal of Experimental Psychology: Gen-
eral, 145, 220 –245. http://dx.doi.org/10.1037/xge0000119
Johanson, C. E., Frey, K. A., Lundahl, L. H., Keenan, P., Lockhart, N.,
Roll, J.,...Schuster, C. R. (2006). Cognitive function and nigrostriatal
markers in abstinent methamphetamine abusers. Psychopharmacology,
185, 327–338. http://dx.doi.org/10.1007/s00213-006-0330-6
Johnson, B. A., Ait-Daoud, N., & Wells, L. T. (2000). Effects of
isradipine, a dihydropyridine-class calcium channel antagonist, on
D-methamphetamine-induced cognitive and physiological changes in
humans. Neuropsychopharmacology, 22, 504 –512. http://dx.doi.org/
10.1016/S0893-133X(99)00116-5
Jordan, C. J., & Andersen, S. L. (2017). Sensitive periods of substance
abuse: Early risk for the transition to dependence. Developmental Cog-
nitive Neuroscience, 25, 29 – 44. http://dx.doi.org/10.1016/j.dcn.2016.10
.004
Kalechstein, A. D., Newton, T. F., & Green, M. (2003). Methamphetamine
dependence is associated with neurocognitive impairment in the initial
phases of abstinence. The Journal of Neuropsychiatry and Clinical
Neurosciences, 15, 215–220.
Khoshbouei, H., Wang, H., Lechleiter, J. D., Javitch, J. A., & Galli, A. (2003).
Amphetamine-induced dopamine efflux. A voltage-sensitive and intracel-
lular Na&-dependent mechanism. The Journal of Biological Chemistry,
278, 12070 –12077. http://dx.doi.org/10.1074/jbc.M212815200
Kim, Y. T., Lee, S. W., Kwon, D. H., Seo, J. H., Ahn, B. C., & Lee, J.
(2009). Dose-dependent frontal hypometabolism on FDG-PET in meth-
amphetamine abusers. Journal of Psychiatric Research, 43, 1166 –1170.
http://dx.doi.org/10.1016/j.jpsychires.2009.03.011
Kopec, J. A., & Esdaile, J. M. (1990). Bias in case-control studies. A
review. Journal of Epidemiology and Community Health, 44, 179 –186.
http://dx.doi.org/10.1136/jech.44.3.179
Linder, J. C., Young, S. J., & Groves, P. M. (1995). Electron microscopic
evidence for neurotoxicity in the basal ganglia. Neurochemistry Interna-
tional, 26, 195–202. http://dx.doi.org/10.1016/0197-0186(94)00110-G
London, E. D., Berman, S. M., Voytek, B., Simon, S. L., Mandelkern,
M. A., Monterosso, J.,...Ling, W. (2005). Cerebral metabolic dys-
function and impaired vigilance in recently abstinent methamphetamine
abusers. Biological Psychiatry, 58, 770 –778. http://dx.doi.org/10.1016/
j.biopsych.2005.04.039
London, E. D., Simon, S. L., Berman, S. M., Mandelkern, M. A., Lichtman,
A. M., Bramen, J.,...Ling, W. (2004). Mood disturbances and regional
cerebral metabolic abnormalities in recently abstinent methamphetamine
abusers. Archives of General Psychiatry, 61, 73– 84. http://dx.doi.org/
10.1001/archpsyc.61.1.73
Loughman, A., Bowden, S. C., & D’Souza, W. (2014). Cognitive func-
tioning in idiopathic generalised epilepsies: A systematic review and
meta-analysis. Neuroscience and Biobehavioral Reviews, 43, 20 –34.
http://dx.doi.org/10.1016/j.neubiorev.2014.02.012
Lynskey, M., & Hall, W. (2000). The effects of adolescent cannabis use on
educational attainment: A review. Addiction, 95, 1621–1630. http://dx
.doi.org/10.1046/j.1360-0443.2000.951116213.x
Mahoney, J. J., III, Jackson, B. J., Kalechstein, A. D., De La Garza, R., II, &
Newton, T. F. (2011). Acute, low-dose methamphetamine administration
improves attention/information processing speed and working memory in
methamphetamine-dependent individuals displaying poorer cognitive per-
formance at baseline. Progress in Neuro-Psychopharmacology & Biologi-
cal Psychiatry, 35, 459 – 465. http://dx.doi.org/10.1016/j.pnpbp.2010.11
.034
Mark, K. A., Soghomonian, J. J., & Yamamoto, B. K. (2004). High-dose
methamphetamine acutely activates the striatonigral pathway to increase
striatal glutamate and mediate long-term dopamine toxicity. The Journal of
Neuroscience, 24, 11449 –11456. http://dx.doi.org/10.1523/JNEUROSCI
.3597-04.2004
McCann, U. D., Kuwabara, H., Kumar, A., Palermo, M., Abbey, R., Brasic,
J., . . . Ricaurte, G. A. (2008). Persistent cognitive and dopamine
transporter deficits in abstinent methamphetamine users. Synapse, 62,
91–100. http://dx.doi.org/10.1002/syn.20471
McCann, U. D., Wong, D. F., Yokoi, F., Villemagne, V., Dannals, R. F.,
& Ricaurte, G. A. (1998). Reduced striatal dopamine transporter density
in abstinent methamphetamine and methcathinone users: Evidence from
positron emission tomography studies with [11C]WIN-35,428. The
Journal of Neuroscience, 18, 8417– 8422. http://dx.doi.org/10.1523/
JNEUROSCI.18-20-08417.1998
McGrew, K. S. (2009). CHC theory and the human cognitive abilities
project: Standing on the shoulders of the giants of psychometric intel-
ligence research. Intelligence, 37, 1–10. http://dx.doi.org/10.1016/j.intell
.2008.08.004
McHugh, M. L. (2012). Interrater reliability: The kappa statistic. Bio-
chemia Medica, 22, 276 –282. http://dx.doi.org/10.11613/BM.2012.031
Meredith, C. W., Jaffe, C., Ang-Lee, K., & Saxon, A. J. (2005). Implica-
tions of chronic methamphetamine use: A literature review. Harvard
Review of Psychiatry, 13, 141–154. http://dx.doi.org/10.1080/106732
20591003605
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & the PRISMA Group.
(2009). Preferred reporting items for systematic reviews and meta-
analyses: The PRISMA statement. PLoS Medicine, 6, 1– 6. http://dx.doi
.org/10.1371/journal.pmed.1000097
Monterosso, J. R., Ainslie, G., Xu, J., Cordova, X., Domier, C. P., & London,
E. D. (2007). Frontoparietal cortical activity of methamphetamine-
dependent and comparison subjects performing a delay discounting task.
Human Brain Mapping, 28, 383–393. http://dx.doi.org/10.1002/hbm.20281
Moon, M., Do, K. S., Park, J., & Kim, D. (2007). Memory impairment in
methamphetamine dependent patients. International Journal of Neuro-
science, 117, 1–9. http://dx.doi.org/10.1080/00207450500535503
Morgan, E. E., Woods, S. P., Poquette, A. J., Vigil, O., Heaton, R. K.,
Grant, I., . . . the Translational Methamphetamine Research Center
(TMARC) group. (2012). Visual memory in methamphetamine-
dependent individuals: Deficient strategic control of encoding and re-
trieval. Australian and New Zealand Journal of Psychiatry, 46, 141–152.
http://dx.doi.org/10.1177/0004867411433212
Morris, L. S., Kundu, P., Dowell, N., Mechelmans, D. J., Favre, P., Irvine,
M. A., . . . Voon, V. (2016). Fronto-striatal organization: Defining
functional and microstructural substrates of behavioural flexibility. Cor-
tex, 74, 118 –133. http://dx.doi.org/10.1016/j.cortex.2015.11.004
Newton, T. F., Kalechstein, A. D., Duran, S., Vansluis, N., & Ling, W.
(2004). Methamphetamine abstinence syndrome: Preliminary findings.
The American Journal on Addictions, 13, 248 –255. http://dx.doi.org/10
.1080/10550490490459915
Newton, T. F., Kalechstein, A. D., Hardy, D. J., Cook, I. A., Nestor, L.,
Ling, W., & Leuchter, A. F. (2004). Association between quantitative
EEG and neurocognition in methamphetamine-dependent volunteers.
Clinical Neurophysiology, 115, 194 –198. http://dx.doi.org/10.1016/
S1388-2457(03)00314-6
Nordahl, T. E., Salo, R., Possin, K., Gibson, D. R., Flynn, N., Leamon, M.,
. . . Sullivan, E. V. (2002). Low N-acetyl-aspartate and high choline in
the anterior cingulum of recently abstinent methamphetamine-dependent
subjects: A preliminary proton MRS study. Magnetic resonance spec-
troscopy. Psychiatry Research, 116, 43–52. http://dx.doi.org/10.1016/
S0925-4927(02)00088-4
O’Neil, M., Kuczenski, R., Segal, D., Cho, A., Lacan, G., & Melega, W.
(2007). Escalating dose pretreatment induces pharmacodynamic and not
pharmacokinetic tolerance to a subsequent high-dose methamphetamine
binge. Synapse, 60, 465– 473.
Parker, J. D., Summerfeldt, L. J., Hogan, M. J., & Majeski, S. A. (2004).
Emotional intelligence and academic success: Examining the transition
from high school to university. Personality and Individual Differences,
36, 163–172. http://dx.doi.org/10.1016/S0191-8869(03)00076-X
Pase, M. P., & Stough, C. (2014). An evidence-based method for exam-
ining and reporting cognitive processes in nutrition in nutrition research.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
751
METHAMPHETAMINE META-ANALYSIS

Nutrition Research Reviews, 27, 232–241. http://dx.doi.org/10.1017/
S0954422414000158
Paulus, M. P., Hozack, N., Frank, L., Brown, G. G., & Schuckit, M. A.
(2003). Decision making by methamphetamine-dependent subjects is
associated with error-rate-independent decrease in prefrontal and pari-
etal activation. Biological Psychiatry, 53, 65–74. http://dx.doi.org/10
.1016/S0006-3223(02)01442-7
Pennay, A. E., & Lee, N. K. (2011). Putting the call out for more research:
The poor evidence base for treating methamphetamine withdrawal. Drug
and Alcohol Review, 30, 216 –222. http://dx.doi.org/10.1111/j.1465-
3362.2010.00240.x
Perry, E. C. (2014). Inpatient management of acute alcohol withdrawal
syndrome. CNS Drugs, 28, 401– 410. http://dx.doi.org/10.1007/s40263-
014-0163-5
Perry, R. P., Hladkyj, S., Pekrun, R. H., & Pelletier, S. T. (2001). Academic
control and action control in the achievement of college students: A
longitudinal field study. Journal of Educational Psychology, 93, 776 –
789. http://dx.doi.org/10.1037/0022-0663.93.4.776
Pope, H. G., Jr., Gruber, A. J., Hudson, J. I., Huestis, M. A., & Yurgelun-
Todd, D. (2001). Neuropsychological performance in long-term canna-
bis users. Archives of General Psychiatry, 58, 909 –915. http://dx.doi
.org/10.1001/archpsyc.58.10.909
Potvin, S., Pelletier, J., Grot, S., Hébert, C., Barr, A., & Lecomte, T.
(2018). Cognitive deficits in individuals with methamphetamine use
disorder: A meta-analysis. Addictive Behaviors, 80, 154 –160. http://dx
.doi.org/10.1016/j.addbeh.2018.01.021
Pu, C., & Vorhees, C. V. (1993). Developmental dissociation of meth-
amphetamine-induced depletion of dopaminergic terminals and astro-
cyte reaction in rat striatum. Developmental Brin Research, 72, 325–
328. http://dx.doi.org/10.1016/0165-3806(93)90201-K
Rendell, P. G., Mazur, M., & Henry, J. D. (2009). Prospective memory
impairment in former users of methamphetamine. Psychopharmacology,
203, 609 – 616. http://dx.doi.org/10.1007/s00213-008-1408-0
Rippeth, J. D., Heaton, R. K., Carey, C. L., Marcotte, T. D., Moore, D. J.,
Gonzalez, R., . . . the HNRC Group. (2004). Methamphetamine depen-
dence increases risk of neuropsychological impairment in HIV infected
persons. Journal of the International Neuropsychological Society, 10,
1–14. http://dx.doi.org/10.1017/S1355617704101021
Robinson, K. A., & Goodman, S. N. (2013). QUADAS-2: A revised tool
for the quality assessment of diagnostic accuracy studies. Annals of
Internal Medicine, 4, 1–10.
Salo, R., Nordahl, T. E., Galloway, G. P., Moore, C. D., Waters, C., &
Leamon, M. H. (2009). Drug abstinence and cognitive control in
methamphetamine-dependent individuals. Journal of substance abuse
treatment, 37(3), 292–297.
Salo, R., Nordahl, T. E., Natsuaki, Y., Leamon, M. H., Galloway, G. P.,
Waters, C.,...Buonocore, M. H. (2007). Attentional control and brain
metabolite levels in methamphetamine abusers. Biological Psychiatry,
61, 1272–1280. http://dx.doi.org/10.1016/j.biopsych.2006.07.031
Salo, R., Nordahl, T. E., Possin, K., Leamon, M., Gibson, D. R., Galloway,
G. P., . . . Sullivan, E. V. (2002). Preliminary evidence of reduced cognitive
inhibition in methamphetamine-dependent individuals. Psychiatry Re-
search, 111, 65–74. http://dx.doi.org/10.1016/S0165-1781(02)00111-7
Salo, R., Ravizza, S., & Fassbender, C. (2011). Overlapping cognitive patterns
in schizophrenia and methamphetamine dependence. Cognitive and Behav-
ioral Neurology, 24, 187–193. http://dx.doi.org/10.1097/WNN.0b013e
31823fc1d0
Salo, R., Ursu, S., Buonocore, M. H., Leamon, M. H., & Carter, C. (2009).
Impaired prefrontal cortical function and disrupted adaptive cognitive
control in methamphetamine abusers: A functional magnetic resonance
imaging study. Biological psychiatry, 65, 706 –709.
Santos, C., & Olmedo, R. E. (2017). Sedative-Hypnotic Drug Withdrawal
Syndrome: Recognition And Treatment. Emergency Medicine Practice,
19, 1–20.
Scammacca, N., Roberts, G., & Stuebing, K. (2014). Meta-analysis with
complex research designs: Dealing with dependence from multiple mea-
sures and multiple group. Review of Educational Research, 84, 328 –
364. http://dx.doi.org/10.3102/0034654313500826
Schmidt, F. L. (1992). What do data really mean? Research findings, meta-
analysis, and cumulative knowledge in psychology. American Psychologist,
47, 1173–1181. http://dx.doi.org/10.1037/0003-066X.47.10.1173
Schumann, G., Loth, E., Banaschewski, T., Barbot, A., Barker, G., Büchel, C.,...
the IMAGEN consortium. (2010). The IMAGEN study: Reinforcement-
related behaviour in normal brain function and psychopathology. Mo-
lecular Psychiatry, 15, 1128 –1139. http://dx.doi.org/10.1038/mp.2010.4
Scott, J. C., Woods, S. P., Matt, G. E., Meyer, R. A., Heaton, R. K.,
Atkinson, J. H., & Grant, I. (2007). Neurocognitive effects of metham-
phetamine: A critical review and meta-analysis. Neuropsychology Re-
view, 17, 275–297. http://dx.doi.org/10.1007/s11065-007-9031-0
Sekine, Y., Minabe, Y., Kawai, M., Suzuki, K., Iyo, M., Isoda, H.,...
Mori, N. (2002). Metabolite alterations in basal ganglia associated with
methamphetamine-related psychiatric symptoms. A proton MRS study.
Neuropsychopharmacology, 27, 453– 461. http://dx.doi.org/10.1016/
S0893-133X(02)00321-4
Sekine, Y., Minabe, Y., Ouchi, Y., Takei, N., Iyo, M., Nakamura, K.,...
Mori, N. (2003). Association of dopamine transporter loss in the orbito-
frontal and dorsolateral prefrontal cortices with methamphetamine-
related psychiatric symptoms. The American Journal of Psychiatry, 160,
1699 –1701. http://dx.doi.org/10.1176/appi.ajp.160.9.1699
Sevak, R. J., Stoops, W. W., Hays, L. R., & Rush, C. R. (2009). Discrim-
inative stimulus and subject-rated effects of methamphetamine,
d-amphetamine, methylphenidate, and triazolam in methamphetamine-
trained humans. The Journal of Pharmacology and Experimental Ther-
apeutics, 328, 1007–1018. http://dx.doi.org/10.1124/jpet.108.147124
Simon, S. L., Dean, A. C., Cordova, X., Monterosso, J. R., & London, E. D.
(2010). Methamphetamine dependence and neuropsychological functioning:
Evaluating change during early abstinence. Journal of Studies on Alcohol and
Drugs, 71, 335–344. http://dx.doi.org/10.15288/jsad.2010.71.335
Sterne, J. A., Gavaghan, D., & Egger, M. (2000). Publication and related
bias in meta-analysis: Power of statistical tests and prevalence in the
literature. Journal of Clinical Epidemiology, 53, 1119 –1129. http://dx
.doi.org/10.1016/S0895-4356(00)00242-0
Tarter, R. E., Kirisci, L., Habeych, M., Reynolds, M., & Vanyukov, M.
(2004). Neurobehavior disinhibition in childhood predisposes boys to
substance use disorder by young adulthood: Direct and mediated etio-
logic pathways. Drug and Alcohol Dependence, 73, 121–132. http://dx
.doi.org/10.1016/j.drugalcdep.2003.07.004
Tarter, R. E., Kirisci, L., Mezzich, A., Cornelius, J. R., Pajer, K., Vanyu-
kov, M.,...Clark, D. (2003). Neurobehavioral disinhibition in child-
hood predicts early age at onset of substance use disorder. The American
Journal of Psychiatry, 160, 1078 –1085. http://dx.doi.org/10.1176/appi
.ajp.160.6.1078
Taylor, M. J., Alhassoon, O. M., Schweinsburg, B. C., Videen, J. S., Grant, I.,
& the HIV Neurobehavioral Research Center. (2000). MR spectroscopy
in HIV and stimulant dependence. Journal of the International Neuro-
psychological Society, 6, 83– 85. http://dx.doi.org/10.1017/S13556
17700611104
Tekin, S., & Cummings, J. L. (2002). Frontal-subcortical neuronal circuits and
clinical neuropsychiatry: An update. Journal of Psychosomatic Research,
53, 647– 654. http://dx.doi.org/10.1016/S0022-3999(02)00428-2
Thompson, P. M., Hayashi, K. M., Simon, S. L., Geaga, J. A., Hong, M. S.,
Sui, Y.,...London, E. D. (2004). Structural abnormalities in the brains of
human subjects who use methamphetamine. The Journal of Neuroscience,
24, 6028 – 6036. http://dx.doi.org/10.1523/JNEUROSCI.0713-04.2004
Villemagne, V., Yuan, J., Wong, D. F., Dannals, R. F., Hatzidimitriou, G.,
Mathews, W. B., . . . Ricaurte, G. A. (1998). Brain dopamine neurotox-
icity in baboons treated with doses of methamphetamine comparable to
those recreationally abused by humans: Evidence from [11C]WIN-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
752 BASTERFIELD, HESTER, AND BOWDEN

35,428 positron emission tomography studies and direct in vitro deter-
minations. The Journal of Neuroscience, 18, 419 –27. http://dx.doi.org/
10.1523/JNEUROSCI.18-01-00419.1998
Volkow, N. (2013). Letter from the director. Retrieved from https://www
.drugabuse.gov/publications/research-reports/methamphetamine/letter-
director
Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Leonido-Yee, M.,
Franceschi, D.,...Miller, E. N. (2001). Association of dopamine
transporter reduction with psychomotor impairment in methamphet-
amine abusers. The American Journal of Psychiatry, 158, 377–382.
http://dx.doi.org/10.1176/appi.ajp.158.3.377
White, H. R., McMorris, B. J., Catalano, R. F., Fleming, C. B., Haggerty,
K. P., & Abbott, R. D. (2006). Increases in alcohol and marijuana use
during the transition out of high school into emerging adulthood: The
effects of leaving home, going to college, and high school protective
factors. Journal of Studies on Alcohol, 67, 810 – 822. http://dx.doi.org/
10.15288/jsa.2006.67.810
Wolf, I. (1986). Meta-analysis: Quantitative methods for research synthesis.
Thousand Oaks, CA: Sage.
Woods, S. P., Rippeth, J. D., Conover, E., Gongvatana, A., Gonzalez, R.,
Carey, C. L., . . . the HIV Neurobehavioral Research Center Group.
(2005). Deficient strategic control of verbal encoding and retrieval in
individuals with methamphetamine dependence. Neuropsychology, 19,
35– 43. http://dx.doi.org/10.1037/0894-4105.19.1.35
Zhong, N., Jiang, H., Du, J., Zhao, Y., Sun, H., Xu, D.,...Zhao, M.
(2016). The cognitive impairments and psychological wellbeing of
methamphetamine dependent patients compared with health controls.
Progress in Neuro-Psychopharmacology & Biological Psychiatry, 69,
31–37. http://dx.doi.org/10.1016/j.pnpbp.2016.04.005
Zweben, J. E., Cohen, J. B., Christian, D., Galloway, G. P., Salinardi, M.,
Parent, D.,...Methamphetamine Treatment Project. (2004). Psychiatric
symptoms in methamphetamine users. The American Journal on Addic-
tions, 13, 181–90. http://dx.doi.org/10.1080/10550490490436055
Received October 22, 2018
Revision received February 28, 2019
Accepted March 1, 2019 !
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
753
METHAMPHETAMINE META-ANALYSIS