JOURNAL OF CHILD AND ADOLESCENT PSYCHOPHARMACOLOGY
Volume 17, Number 5, 2007
© Mary Ann Liebert, Inc.
Methylphenidate Effects on Functional Outcomes in the
Preschoolers with Attention-Deficit/Hyperactivity
Disorder Treatment Study (PATS)
Howard B. Abikoff, Ph.D.,1Benedetto Vitiello, M.D., M.S.,2Mark A. Riddle, M.D.,3
Charles Cunningham, Ph.D.,4Laurence L. Greenhill, M.D.,5James M. Swanson, Ph.D.,6
Shirley Z. Chuang, M.S.;7Mark Davies, M.P.H.,5Elizabeth Kastelic, M.D.,3
Sharon B. Wigal, Ph.D.,6Lori Evans, Ph.D.;1Jaswinder K. Ghuman, M.D.,8M.D.,
Scott H. Kollins, Ph.D.,9James T. McCracken, M.D.,10James J. McGough, M.D.,10
Desiree W. Murray, Ph.D.,9Kelly Posner, Ph.D.,5Anne M. Skrobala, M.A.,5
and Tim Wigal, Ph.D.6
Objective: The purpose of this study was to examine the effects of methylphenidate (MPH)
on functional outcomes, including children’s social skills, classroom behavior, emotional sta-
tus, and parenting stress, during the 4-week, double-blind placebo controlled phase of the
Preschoolers with Attention Deficit/Hyperactivity Disorder (ADHD) Treatment Study (PATS).
Methods: A total of 114 preschoolers who had improved with acute MPH treatment, were
randomized to their best MPH dose (M ? 14.22 mg/day; n ? 63) or placebo (PL; n ? 51). As-
1New York University Child Study Center, New York, New York.
2National Institute of Mental Health,
3Johns Hopkins University, Baltimore, Maryland.
4McMaster University, Hamilton, Ontario.
5New York State Psychiatric Institute/Columbia University, New York, New York.
6University of California Irvine, Irvine, California.
7No current affiliation, formerly at New York State Psychiatric Institute/Columbia University.
8University of Arizona, Tucson, Arizona.
9Duke University Medical Center; Durham, North Carolina.
10University of California Los Angeles, Los Angeles, California.
Statistical consultants: Shirley Z. Chuang, M.S., Mark Davies, M.P.H. Ms. Chuang has no current affiliation; she
was formerly at New York State Psychiatric Institute/Columbia University. Mr. Davies is affiliated with New York
State Psychiatric Institute/Columbia University.
This research was supported by a cooperative agreement grants between the National Institute of Mental Health
and the following institutions: Johns Hopkins University (U01 MH60642), NYSPI/Columbia University (U01
MH60903), University of California, Irvine (U01 MH60833), Duke University Medical Center (U01 MH60848), New
York University Child Study Center (U01 MH60943), and University of California, Los Angeles (U01 MH60900); and
by NIMH K23 MH01883-01A1 and Arizona Institute of Mental Health Research to J.K.G.
The opinions and assertions contained in this report are the private views of the authors and are not to be con-
strued as official or as reflecting the views of the Department of Health and Human Services, the National Institutes
of Health, or the National Institute of Mental Health.
made in middle childhood, although onset dur-
ing preschool years is typical. In contrast to an
extensive treatment literature regarding the
efficacy of psychostimulants in elementary
school-aged children with ADHD, only a small
number of such studies have been conducted
with preschoolers. The efficacy of psychostim-
ulants in reducing ADHD symptoms in
preschoolers was first reported in the 1970s
(Conners 1975; Schleifer et al. 1975). With few
exceptions (Cohen 1981; Barkley et al. 1984),
subsequent placebo-controlled studies, al-
though varying in design, quality, and size,
have confirmed medication effects on symp-
toms of ADHD. Monteiro-Musten et al. (1997)
found that stimulants increased preschoolers’
attention and decreased impulsiveness. Byrne
et al. (1998) reported that stimulants improved
behavior and significantly reduced errors of
omission on visual and auditory vigilance tests.
Short et al. (2004) found a clinically significant
reduction in ADHD symptoms in 82% (n ? 28)
of preschoolers treated with stimulants. Find-
ings from the National Institute of Mental
Health (NIMH) multisite Preschoolers with
ADHD Treatment Study (PATS) indicated that
immediate release methylphenidate (MPH-IR),
delivered in 2.5-, 5.0-, and 7.5-mg doses three
times a day (t.i.d.), yielded significant reduc-
tions on ADHD symptom scales compared to
placebo (Greenhill et al. 2006).
School-aged and preschool children with
HE DIAGNOSIS OF attention-deficit/hyperac-
tivity disorder (ADHD) is most commonly
ADHD share not only a common symptom
profile, but also a similar pattern of associated
functional deficits and impairments (Sonuga-
Barke et al. 2003). Both age groups have deficits
in social skills, especially in social cooperation
(Merrell and Wolfe 1998) and friendships (La-
hey et al. 1998). They also experience problem-
atic interactions with their parents and other
relatives (DuPaul et al. 2001; Daley et al. 2003)
that contribute to high levels of familial stress,
which in turn exacerbate mental health prob-
lems among family members (DeWolfe et al.
Notably, functional impairments, especially
in social functioning and parent–child interac-
tions, as well as deficient regulation of emo-
tions and problematic classroom behavior, typ-
ically result in clinic referrals in children with
ADHD. Medication has been shown to im-
prove these functional domains in elementary
school-aged children with ADHD (MTA Co-
operative Group 1999; Abikoff et al. 2004;
Hechtman et al. 2004); however, information
regarding medication effects on these aspects
of functioning in preschoolers with ADHD is
sparse and inconclusive. Acute MPH treatment
has been reported to reduce the observed fre-
quency of controlling and dominating peer in-
teractions in 4–6 year olds with ADHD in a sim-
ulated classroom setting (Cunningham et al.
1985). Barkley (1988) reported that stimulants
improved the quality of interactions between
preschoolers and their mothers, whereas Mon-
teiro-Musten et al. (1997) found that, although
stimulants improved adjustment, they did not
increase compliance with parental requests.
ABIKOFF ET AL.582
sessments included the Clinical Global Impression-Severity (CGI-S), parent and teacher ver-
sions of the Strengths and Weaknesses of ADHD-Symptoms and Normal Behaviors (SWAN),
Social Competence Scale (SCS), Social Skills Rating System (SSRS), and Early Childhood In-
ventory (ECI), and Parenting Stress Index (PSI).
Results: Medication effects varied by informant and outcome measure. Parent measures and
teacher SWAN scores did not differentially improve with MPH. Parent-rated depression (p ?
0.02) and dysthymia (p ? 0.001) on the ECI worsened with MPH, but scores were not in the
clinical range. Significant medication effects were found on clinician CGI-S (p ? 0.0001) and
teacher social competence ratings (SCS, p ? 0.03).
Conclusions: Preschoolers with ADHD treated with MPH for 4 weeks improve in some as-
pects of functioning. Additional improvements might require longer treatment, higher doses,
and/or intensive behavioral treatment in combination with medication.
In recognition of the need to characterize the
impact of stimulants on clinically relevant
aspects of functioning in preschoolers with
ADHD, the PATS trial evaluated the effects of
MPH on children’s social skills, classroom be-
havior, emotional status, and parenting stress.
We hypothesized that medication effects in
preschoolers with ADHD would parallel those
found in school-aged children. Specifically, it
was hypothesized that: (1) compared to chil-
dren randomized to placebo (PL), preschoolers
randomized to their “best dose” of MPH would
show significant improvements in classroom
behavior, social, and emotional functioning;
and (2) parents of children receiving MPH
would show significant reduction in parenting
stress compared to parents of children on PL.
Detailed descriptions of the PATS design and
methods are provided elsewhere (Greenhill et
al. 2006; Kollins et al. 2006); therefore, only key
study features are presented here.
Participants had to meet the following crite-
ria: Stimulant-naive children of both sexes, ages
3–5.5 years with a Diagnostic and Statistical Man-
ual of Mental Disorders, Fourth Edition (DSM-
IV) consensus diagnosis of ADHD (American
Psychiatric Association, 1994), combined or
predominantly hyperactive subtype, based on
the Diagnostic Interview Schedule for Children
IV–Parent Version (DISC-IV-P) (Shaffer et al.
1996) and semistructured interview; an im-
pairment scale score ?55 on the Children’s
Global Assessment Scale (C-GAS) (Shaffer et al.
1983); hyperactive-impulsive subscale T score
of 65 (1.5 SD above age- and sex-adjusted
means) on both the Conners Revised Parent
(CPRS) (Conners et al. 1998a) and Teacher
(CTRS) (Conners et al. 1998b) Rating Scales; full
scale intelligence quotient (IQ) ? 70 on the Dif-
ferential Ability Scales (DAS; Elliott 1990); par-
ticipation in a preschool, day-care group set-
ting or other school program at least 2 half-days
per week with at least 8 same-age peers; and
the same primary caretaker for at least 6
months prior to screening.
Exclusion criteria included current evidence
of adjustment disorder, pervasive develop-
mental disorders, psychosis, significant suici-
dality, or other psychiatric disorder that re-
quired treatment with additional medication;
current stimulant or cocaine abuse in a relative
living in the home; a confounding medical con-
dition; inability of the parent to understand or
follow study instructions; or history of bipolar
disorder in both biological parents.
The PATS design consisted of several phases,
detailed in Kollins et al. (2006). Briefly, 183
preschoolers who failed to improve signifi-
cantly after parent participation in a 10-week
parent training program, and whose parents
consented to a medication trial for their young-
sters, entered an open-label, lead-in phase to
determine if they could tolerate doses of MPH
used in the subsequent double-blind phases.
Children (n ? 165) who tolerated the lead-in
doses entered the study’s 5-week double-blind,
within-subject, placebo-controlled, crossover-
design, titration phase, and 147 children were
randomized to, and completed, all five se-
quences of active MPH (1.25, 2.5, 5, 7.5 mg) and
PL administered t.i.d.. At the end of each week,
school and home behavior and side effects rat-
ings were obtained from teachers and parents.
At the conclusion of the 5-week trial, these
weekly ratings were reviewed by two inde-
pendent clinicians, blind to dosing information,
who generated a consensus decision regarding
each child’s best dose. Following this cross-
over phase, children went on to participate in
the between-subjects, randomized, 4-week,
double-blind, parallel design study phase of
the PATS trial, wherein they were randomized
to either their best dose of MPH or PL. The par-
allel-design phase, unlike the preceding titra-
tion phase, included assessments of functional
outcomes. Findings on these outcomes are re-
The study protocol and parental informed
consent forms were approved by the institu-
tional review boards at the six recruiting sites.
The study was monitored by the Data and
Safety Monitoring Board of the National Insti-
tute of Mental Health.
FUNCTIONAL OUTCOMES IN THE PATS 583
A total of 114 children participated in the
parallel design study phase of the PATS: 61
were randomized to MPH and 53 to PL. Their
characteristics are summarized in Table 1.
Domains and measures
ADHD behavior. Strengths and Weaknesses
of ADHD-Symptoms and Normal Behaviors.
The common occurrence of ADHD behaviors
in preschoolers results in over-estimates of
such behaviors with conventional ADHD rat-
ing scales. The SWAN’s 7-point metric (from
?3 “far above average” to ?3 “far below av-
erage”) of strengths and weaknesses of ADHD-
Symptoms and Normal Behaviors (SWAN)
was designed to protect against over-estimates
of ADHD behaviors, and yields ratings of
preschool children’s ADHD behaviors that are
normally distributed, such that average, unaf-
fected children would receive ratings of “0,”
which indicates no particular attentional or im-
pulsive/motor problems compared to other
children of the same age. Teacher and parent
versions of the SWAN were used in the study
(Swanson et al. 2001; Cornish et al. 2005).
Social behavior. Social Skills Rating System.
The Social Skills Rating System (SSRS) is a stan-
dardized scale that assesses social functioning.
Preschool and School-age parent (SSRS-P) and
teacher (SSRS-T) versions are available (Gre-
sham and Elliott 1990). The SSRS-P comprises
70 items, 60 of which assess prosocial skills and
10 assess problem behaviors. Fifty of the proso-
ABIKOFF ET AL.584
TABLE 1.DEMOGRAPHIC AND CLINICAL CHARACTERISTICS OF THE SAMPLE AT BASELINE
Variable Best dose (n ? 61) Placebo (n ? 53)
Age, years mean (SD)
Male, n (%)
Ethnicity, n (%)
Black or African American
Hispanic or Latino
High school graduate, n (%)
Employed, n (%)
Welfare, n (%)
Married, n (%)
Hollingshead SES, mean (SD)
Family Composition, n (%)
CTRS, mean (SD)
CPRS, mean (SD)
DAS IQ, mean (SD)
C-GAS Impairment Scale, mean (SD)
ADHD Subtype, n (%)
CTRS ? Conners Teacher Rating Scale; CPRS ? Conners Parent Rating Scale; DAS ? Dif-
ferential Ability Scale; C-GAS ? Children’s Global Assessment Scale; SD ? standard de-
viation; IQ ? intelligence quotient; SES ? socio-economic status.
ap ? 0.04, all other comparisons nonsignificant.
cial items overlap with the SSRS-T and the
other 10 address social situations at home. The
Total Social Skills score served as the study out-
come measure, with higher scores indicating
better functioning. Reliability and validity have
been established (Gresham and Elliott, 1990).
Social Competence Scale. The 12-item parent
(SCS-P) and teacher (SCS-T) versions of the So-
cial Competence Scale Scale assess frustration
tolerance, peer relationships, communication
skills, and empathy (Conduct Problems Pre-
vention Research Group 1992). The Total score
served as the study outcome measure, with
higher scores indicating more competence. The
scale has demonstrated sensitivity to treatment
effects in preschoolers (Webster-Stratton 1998).
Parental stress. Parenting Stress Index. The
Parenting Stress Index (PSI) is a 101-item, 5-
point Likert scale that measures stressful child,
parental, relational, and situational character-
istics (Abidin 1995). The Total score on the Par-
ent Domain, which taps depression, attach-
ment, restriction of role, sense of competence,
social isolation, relation with spouse, and par-
ent health, served as the outcome measure. In-
ternal consistency and concurrent and dis-
criminant validity have been documented. The
scale has been shown to be sensitive to treat-
ment effects in children with conduct disorders
(Kazdin and Whitley, 2003).
Mood.Early Child Inventory. The Early Child
Inventory (ECI) is a 108-item inventory that has
been normed on preschool children, and the
parent (ECI-P) and teacher (ECI-T) versions
have acceptable reliability and validity data
(Gadow and Sprafin, 1996; Sprafkin et al. 2002).
The items assess behaviors and symptoms as-
sociated with a wide-range of childhood men-
tal disorders. Scores on the two Mood Scales,
Dysthymic Disorder and Major Depressive
Disorder, served as outcome measures.
Severity of illness. Clinical Global Impres-
sions–Severity. The Clinical Global Impres-
sions–Severity (CGI-S) is a clinician completed
7-point rating scale (1 ? not at all ill to 7 ?
among the most extremely ill) that assesses the
child’s current level of severity of illness (Guy
1976). The scale has adequate psychometric
properties (American Psychiatric Association,
2000) and is one of the most widely used out-
come measures in psychopharmacology trials.
Measures were obtained at baseline at the be-
ginning of the PATS trial (prior to initiation of
medication) and at the end of the 4-week, dou-
ble-blind, parallel design study phase. For par-
ticipants who did not complete the 4-week
phase, efforts were made to obtain outcome
scores at the time of dropout. Data analysis was
based on intent-to-treat, last observation car-
ried forward (LOCF) procedures. Analyses of
covariance of fixed effects, using baseline
scores as the covariate, were conducted to eval-
uate medication effects on the study measures.
Effect sizes (ES), using Cohen’s d, were calcu-
lated for each outcome measure by dividing the
difference between the post-treatment group
means by the pooled SD.
The CGI-S, completed by the clinician at the
end of each subject’s participation in the par-
allel-design phase, was obtained on the full
sample. Data on other measures were missing
from some parents, and to a greater extent,
from teachers. An a priori decision was made
to analyze measures for which scores were
available on at least half of the sample that en-
tered the double-blind (DB) efficacy trial. Con-
sequently, the teacher-completed ECI, which
was unavailable on 60% of the sample (38 in
the MPH group [62.6 %] and 30 in the PL group
[56.6 %]) and the teacher-completed SSRS,
missing for 51% of the sample (31 in the MPH
group [50.8 %] and 27 in the PL group [50.1%]),
were not analyzed. Descriptive statistics, anal-
ysis of covariance (ANCOVA) results and ef-
fect sizes are presented in Table 2.
Of the 114 children who entered the 4-week
parallel-design phase, 36 (32%) dropped out.
Of these, 33 had behavioral deterioration [24
were in the PL group (24/53, 45%), and 9 in the
MPH group (9/61, 15%)], 2 declined study par-
ticipation, and 1 had medication related ad-
FUNCTIONAL OUTCOMES IN THE PATS 585
COMPARISON OF TREATMENT GROUPS ON STUDY OUTCOME MEASURES AT IMMEDIATE POSTTREATMENT
Best dose (n ? 61)
Placebo (n ? 53)
aLower scores indicate better outcome.
bHigher scores indicate better outcome.
SWAN ? Strengths and Weaknesses of ADHD-Symptoms and Normal Behaviors; Hyp-Imp ? Hyperactive/Impulsive scale; ECI ? Early Child Inventory; MDD ? ma-
jor depressive disorder; CGI-S ? Clinical Global Impressions-Severity; PSI ? Parenting Stress Index; SCS ? Social Compentence Scale; SSRS ? Social Skills Rating System;
ADHD ? attention-deficit/hyperactivity disorder; SD ? standard deviation; NS ? not significant; ANCOVA ? analysis of covariants.
verse effects. Nineteen children dropped out in
week 1 (15 on PL, 4 on MPH), 13 in week 2
(PL ? 8, MPH ? 5) and one in week 3 (PL).
Analyses comparing the baseline characteris-
tics of completers and non-completers indi-
cated no significant differences.
On the SWAN, the MPH and PL groups did
not differ significantly in parent or teacher rat-
ings of total ADHD or inattention, or parent
ratings of hyperactivity/impulsivity, with
scores improving in both groups from pre to
post. However, teacher ratings of hyperactiv-
ity/impulsivity indicated a trend (p ? 0.08;
ES ? 0.27) in favor of the MPH group.
Children’s social competence and social
skills, based on parent ratings on the SCS and
SSRS, respectively, did not differ significantly
between the groups. A significant treatment ef-
fect (p ? 0.03; ES ? 0.39) was found in teacher
ratings on the SCS, with preschoolers treated
with medication improving in their social com-
petence scores, whereas those on placebo
showing no change from pre to post.
There was no significant difference in the PSI
ratings of parents of children in the MPH and
PL groups, with the mean Total scores de-
creasing in both groups from pre to post.
Compared to preschoolers treated with PL,
children in the MPH group had significantly
higher mood scores on the parent rated ECI Ma-
jor Depressive Disorder (p ? 0.02) and Dys-
thymic scales (p ? 0.001). The group difference
reflected an increase in mood symptoms from
pre- to posttreatment with MPH in contrast to
a decrease over time with PL (see Table 2). Sub-
sequent item analyses of the mood symptoms
comprising these two ECI scales indicated that
the item “Has become more sensitive or tearful
than usual” was the only symptom that signif-
icantly differentiated MPH from PL (p ? 0.003).
Severity of illness
Clinicians’ global ratings of severity on the
CGI-S at posttreatment were significantly
lower for children in the MPH group compared
to those on PL (p ? 0.0001; ES ? 0.73).
Contrary to expectations, the effects of MPH
on functional outcomes in preschoolers did not
parallel the functional improvements reported
in elementary school-aged youngsters with
ADHD treated with MPH. Rather, in the cur-
rent study, medication effects varied as a func-
tion of informant and outcome measure. The
absence of MPH effects was most evident on
parent measures. Preschoolers’ social skills and
social competence, as well as their level of
ADHD behaviors on the SWAN, did not im-
prove differentially with medication on the ba-
sis of parent ratings. Similarly, self-ratings of
parental stress were not significantly different
in parents of children treated with MPH or PL.
Moreover, parents’ ratings of their children’s
mood symptoms indicated a worsening of
symptoms with medication. In contrast, posi-
tive medication effects were detected in clini-
cians’ global severity ratings and in teachers’
ratings of improved social competence in chil-
dren on MPH. However, like parents, teachers’
ratings of ADHD behaviors on the SWAN were
not differentially related to the children’s med-
A variety of factors need to be considered in
interpreting the study outcomes. Forty five per-
cent of the children in the PL group dropped
out before the end of the 4-week treatment
phase because of behavioral deterioration,
compared to 15% of children receiving MPH.
At dropout, efforts to obtain outcome mea-
sures, which were intended to serve as LOCF
data, were not always successful. Conse-
quently, the power to detect treatment differ-
ences was decreased because of the reduced
sample size available for analysis.
The lack of terminal data on all dropouts pre-
cludes an explication of the exact nature of the
“behavioral deterioration” in these youngsters.
In the PATS primary efficacy paper (Greenhill
FUNCTIONAL OUTCOMES IN THE PATS 587
et al. 2006), we reported that attrition during
the parallel group double-blind phase “was
significantly correlated with elevated SNAP
scores (p ? 0.009)” (p. 1291), indicating that
children withdrawn from the study had an in-
crease in ADHD symptoms. It is unknown if
these dropouts also showed an increase in
functional impairment. To the extent that they
did, the results reported here may be underes-
timates of MPH effects on functional outcomes.
However, it is important to emphasize that the
association between ADHD symptom severity
and degree of functional impairment has been
reported to be relatively small, with symptom
severity predicting less than 25% of the vari-
ance in impairment in four separate studies
(Gordon et al. 2006).
An underestimation of MPH effects could
have also occurred if the completers were less
functionally impaired on placebo than those
who dropped out of the PL group. Here too,
the absence of terminal ratings precludes a di-
rect test of this possibility. However, as an in-
direct test of this notion, we examined if the
completers and dropouts had a differential re-
sponse to placebo during the preceding dou-
ble-blind titration phase. Independent t-tests
indicated no significant differences (p values
ranged from 0.32 to 0.82) in the subgroups’
scores on the parent and teacher Connors,
Loney, and Milich (CLAM) and Swanson,
Kotkin, Atkins, M-Flynn, and Pelhan (SKAMP)
ADHD rating scales during the titration week
when each child was on placebo.
The higher-than-expected rate of premature
treatment discontinuation may be related to the
multistage, sequential design of PATS. Children
participating in the placebo-controlled trial re-
ported here had previously completed a within-
subject titration showing superiority of MPH
over placebo at the individual patient level. Par-
ents’ experience during the titration phase pre-
sumably heightened their awareness of the be-
havioral differences associated with active and
placebo medication. Such knowledge, in con-
junction with study guidelines that allowed
parents to forego or discontinue participation in
the double-blind, parallel-group phase and
have their child move directly to open mainte-
nance treatment with MPH, likely contributed
to dropout decisions for some parents.
We considered the possibility that the
dropouts in the PL group showed more im-
provement with medication during titration
than did the completers in the PL group, in-
creasing the likelihood of dropout. To this end,
we compared these two subgroups’ parent and
teacher scores on the CLAM and SKAMP
ADHD rating scales when they were on their
optimal dose of MPH during titration. The PL
dropouts and completers did not differ signif-
icantly on any of these measures.
The MPH doses in the double-blind parallel
group study phase were relatively low (M ?
14.22 ? 8.1 mg/day), which may have limited
functional improvements Findings from the
PATS maintenance phase (Vitiello et al., this is-
sue), provide some support for this notion. For
example, the increase in children’s MPH doses
from the first to the tenth month of mainte-
nance treatment (M ? 19.98 ? 9.56 mg/day)
was associated with improvements in chil-
dren’s social skills as rated by parents. These
results must be tempered because of the ab-
sence of an untreated group as a temporal con-
trol. However, future studies might explore the
possibility that children’s social functioning is
facilitated with higher doses.
Although MPH resulted in a significant in-
crease in scores on the Major Depressive Dis-
order and Dysthymic Disorder scales on the
ECI compared to PL, the scores were not in the
clinical range. The higher score in the medica-
tion group on the item “Has become more sen-
sitive or tearful than usual” likely reflects the
significant increase in emotional outbursts/
crying with MPH compared to PL reported in
the initial titration phase of the study (Wigal et
al. 2006). Notably, the frequency of emotional
outbursts/crying decreased significantly dur-
ing the 10-month maintenance period, even
though the mean MPH total daily dose in-
creased during this period (Wigal et al. 2006).
A similar pattern of findings has been reported
in 7- to 9-year-old children with ADHD, who
showed initial increases in their Children’s De-
pression Inventory scores after 5 weeks of MPH
treatment, followed by a significant decrease in
CDI scores after 6 months of treatment with
MPH (Hechtman et al. 2004).
Teachers, unlike parents, reported gains in
social competence with MPH. Poor concor-
ABIKOFF ET AL.588
dance rates for parent and teacher ratings of
ADHD symptoms in the PATS sample has been
reported by Murray et al. (this issue). The lack
of agreement between parent and teacher rat-
ings of children’s behavior is well established,
and, in children with ADHD, it is considered
to reflect the variability of children’s behaviors
across situations (McDermott et al. 2005). Re-
latedly, contextual factors could have facili-
tated teachers’ ability to detect medication re-
lated changes in social competence, because
they had more opportunities than parents to
see children in social situations. These infor-
mant differences illustrate the importance of
collecting and analyzing information from key
individuals regarding preschool children’s
functioning to obtain a more complete picture
of treatment outcome.
Stimulant medication in children with
ADHD has been reported to have positive ef-
fects on some, but not all aspects of parenting.
Improved parent–child interactions, decreased
negative parenting practices, and improved
ratings of parental effectiveness often occur
with MPH treatment (Stein et al. 1996; Hecht-
man et al. 2004). In contrast, medication has not
been shown to increase positive parenting
practices (Hechtman et al. 2004) or positive
changes in parent functioning, such as im-
proved mood and ability to complete tasks
(Chronis et al. 2003). In the current study,
parental stress was not differentially reduced
in parents whose children were treated with
MPH relative to those on PL. The reasons for
this are not clear. It is likely that children
showed symptomatic improvement through-
out much of the day. However, parental diffi-
culties in managing their children in the morn-
ing, before medication effects were observable,
or later in the day, when medication effects dis-
sipated, conceivably continued to be salient,
stressful events that influenced parents’ judg-
ments and self-ratings of stress levels.
Clinicians’ global impressions of illness
severity were significantly reduced in children
treated with MPH compared to those on PL,
yielding an ES of 0.73, the largest obtained on
any outcome measure. CGI-S ratings, which
take into account the youngster’s overall func-
tioning, have been reported to be influenced by
a variety of factors, including severity of
ADHD symptoms, peer relationship problems,
oppositional defiant disorder (ODD), conduct
disorder (CD), and internalizing symptoms
(Coghill et al. 2006). Consequently, the degree
to which clinicians’ CGI-S ratings were influ-
enced by functional improvements and/or re-
ductions in children’s ADHD symptoms is un-
known. However, the absence of treatment
effects in teacher and parent SWAN ratings of
ADHD symptoms, with children in both
groups improving over time, suggests that clin-
icians’ judgments reflected, at least in part, im-
provements in non-ADHD-related areas of
The present study has some limitations. First,
as described above, the differential attrition
rates in the two groups limited power to detect
treatment differences and may have resulted in
an underestimation of MPH effects on func-
tional outcomes. Second, missing data, espe-
cially from teachers, precluded analyses of
teacher ratings on the SSRS and ECI. As a re-
sult, it is unknown whether broader aspects of
children’s social skills, beyond those associated
with social competence, improved in the school
setting with MPH. Similarly, it is unknown
whether the increase in sensitivity and tearful-
ness with MPH reported by parents occurred
in school as well. Third, little is known about
the psychometric properties of the CGI-S in
preschool-aged children. The absence of relia-
bility and validity information in this age
group needs to be considered in interpreting
the findings on the CGI-S. Fourth, in an effort
to minimize treating preschoolers with med-
ication unnecessarily, the PATS study design
incorporated several features, including a 10-
week parenting program, which preceded the
study’s medication phases, and a conservative
set of diagnostic and inclusion criteria to min-
imize false positive diagnoses. These criteria in-
cluded elevated scores on both the teacher and
parent Conners Rating scales and an impair-
ment score on the C-GAS ?55. The resulting
study sample had a mean C-GAS score of 47,
which is considerably lower than the typical C-
GAS score of 65 for children with ADHD seen
in clinic settings. Consequently, the generaliz-
FUNCTIONAL OUTCOMES IN THE PATS589
ability of the findings reported here to
preschoolers with less severe levels of ADHD
and impairment is uncertain.
Finally, although 4 weeks is typically suffi-
cient to detect improvements in ADHD symp-
toms, a longer treatment period may be needed
for functional changes to occur. However, be-
cause of design features of the PATS trial, it
was not deemed ethical or clinically viable to
keep children on placebo for more than 4
weeks. Specifically, the participants in the par-
allel-group phase had all participated in the
previous within-subject titration phase and had
demonstrated benefit with MPH treatment. In
light of their prior experience with MPH, there
was concern about the length of time individ-
uals would tolerate treatment with PL, which
led to the decision to limit the randomized, par-
allel-group phase to four weeks. Even with this
design, the dropout rate was quite high in the
PL group. It is likely that attrition in the PL
group would have been even higher with an
extended randomized, parallel-group phase.
However, a parallel-group design that in-
cluded randomization to MPH or PL for a
longer treatment period, such as 8 weeks,
might be viable in preschoolers without any
prior exposure to MPH. Such an extended
treatment period would provide greater op-
portunity for any functional improvements
that occur to consolidate and be detected.
Preschoolers with ADHD treated with MPH
for 4 weeks show some improvements in func-
tioning, although not as extensive as those
found in their elementary school-aged counter-
parts. We cannot rule out the possibility that
these different outcome patterns are an artifact
of the study design and uneven attrition rates
in the MPH and PL groups. Nonetheless, it may
be that additional functional improvements in
preschoolers with ADHD require a longer treat-
ment period, higher doses than those used in
the current study, and/or the use of intensive
behavioral treatment in combination with med-
ication. Although there is a scarcity of evidence-
based psychosocial interventions for preschool
children with ADHD, the inclusion of “estab-
lished” parenting programs [e.g., New Forest
Parenting Program (Sonuga-Barke et al. 2001);
Parent-Child Interaction Therapy (Eyberg et al.
1995)], perhaps in conjunction with school-
based behavioral approaches, might lead to ad-
ditional improvements in children’s social func-
tioning and in parent stress levels beyond those
reported here. Systematic, randomized con-
trolled trials are needed to address this issue.
The following financial disclosures indicate
potential conflicts of interest among the PATS
investigators and industry sources for the
period 2000–2007. 1Honoraria/consultant, 2re-
search support, 3speaker’s bureau, 4significant
equity (?$50,000). Dr. Abikoff: Abbott Labora-
tra-Zeneca,1Pfizer.2Dr. Swanson: Alza,1,2Cel-
Novartis,1,2UCB, Shire.1,2Dr. Greenhill: Cell-
Therapeutics Communications.1Dr. Kollins:
agen,1,2Cephalon.1Dr. McCracken: Abbott,
UCB, Novartis, Johnson & Johnson, Eli
Noven,1Bristol Meyers Squibb,1Janssen,1,3
Wyeth.1Dr. McGough: Eli Lilly,1,2,3McNeil,1,2,3
Dr. Murray: Eli Lilly,2
Pfizer,2Dr. Posner: As part of an effort to help
execute the FDA suicidality classification man-
dates, Dr. Posner has had research support (2)
from GlaxoSmithKline, Forest Laboratories, Ei-
sai Inc., Astra Zeneca Pharmaceuticals, Johnson
and Johnson, Abbott Laboratories, Wyeth Re-
search, Organon USA, Bristol Meyers Squibb,
Sanofi-Aventis, Cephalon, Novartis, Shire
Pharmaceuticals and UCB Pharma; Shire.1,2
Dr. Sharon Wigal: Celltech,1,2,3McNeil,1,2,3
ABIKOFF ET AL. 590
River Pharmaceuticals,2Janssen,3Eli Lilly.2Dr.
Ghuman: Bristol Myers-Squibb.2Dr. Tim Wi-
gal: Celltech,2Cephalon,2Eli Lilly,2,3Janssen,2
Cunningham, Kastelic, Evans, Ms. Chuang,
and Ms. Skrobala have no conflict of interest as
they have not received support from compa-
nies manufacturing medications.
Abidin RR: Parenting Stress Index, 3rded. Lutz (Fl), Psy-
chological Assessment Resource, 1995.
Abikoff H, Hechtman L, Klein R, Gallagher R, Fleiss K,
Etcovitch J, Cousins L, Greenfield B, Martin D, Pollack
S: Social functioning in children with ADHD treated
with long-term methylphenidate and multimodal psy-
chosocial treatment. J Am Acad Child Adolesc Psychi-
atry 43:812–819, 2004.
American Psychiatric Association. Diagnostic and Statis-
tical Manual of Mental Disorders, 4thedition (DSM-IV).
Washington (DC), American Psychiatric Association,
American Psychiatric Association: Handbook of Psychi-
atric Measures. Washington (DC), American Psychi-
atric Association, 2000.
Barkley RA: The effects of methylphenidate on the interac-
tions of preschool ADHD children with their mothers. J
Am Acad Child Adolesc Psychiatry 27:336–341, 1988.
Barkley RA, Karlsson J, Strzelecki E, Murphy JV: Effects
of age and Ritalin dosage on the mother-child interac-
tions of hyperactive children. J Consult Clin Psychol
Byrne JM, Bawden HN, DeWolfe NA, Beattie TL: Clini-
cal assessment of psychopharmacological treatment of
preschoolers with ADHD. J Clin Exp Neuropsychol
Chronis AM, Pelham WE Jr, Gnagy EM, Roberts JE,
Aronoff HR: The impact of late-afternoon stimulant dos-
ing for children with ADHD on parent and parent-child
domains. J Clin Child Adolesc Psychol 32:118–126, 2003.
Coghill D, Spiel G, Baldursson G, Dopfner M, Lorenzo
MJ, Ralston SJ, Rothenberger A, ADORE Study Group:
Which factors impact on clinician-rated impairment in
children with ADHD? Eur Child Adolesc Psychiatry
15 (Suppl 1):1/30–1/37, 2006.
Cohen NJ: Evaluation of the relative effectiveness of
methylphenidate and cognitive behavior modification
in the treatment of kindergarten-aged hyperactive
children. J Abnorm Child Psychol 9:43–54, 1981.
Conduct Problems Prevention Research Group: A devel-
opmental and clinical model for the prevention of con-
duct disorders: The Fast Track program. Dev Psy-
chopathol 4: 509–527, 1992.
Conners CK: Controlled trial of methylphenidate in
preschool children with minimal brain dysfunction. Int
J Ment Health 4:61-74, 1975.
Conners CK, Sitarenios G, Parker JD, Epstein JN: The re-
vised Conners’ Parent Rating Scale (CPRS-R): Factor
structure, reliability, and criterion validity. J Abnorm
Child Psychol 26 257–268, 1998a.
Conners CK, Sitarenios G, Parker JDA, Epstein JN: Revi-
sion and restandardization of the Conners’ Teacher
Rating Scale (CTRS-R): Factor structure, reliability, and
criterion validity. J Abnorm Child Psychol 26:279–291,
Cornish KM, Manly T, Savage R, Swanson J, Morisano D,
Butler N, Grant C, Cross G, Bentley L, Hollis CP: As-
sociation of the dopamine transporter (DAT1) 10/10-
repeat genotype with ADHD symptoms and response
inhibition in a general population sample. Mol Psy-
chiatry 10:686–698, 2005.
Cunningham CE, Siegel LS, Offord DR: A developmen-
tal dose response analysis of the effects of methyl-
phenidate on the peer interactions of attention deficit
disordered boys. J Child Psychol Psychiatry 26:955–
Daley D, Sonuga-Barke EJS, Thompson M: Assessing ex-
pressed emotion in mothers of preschool AD/HD chil-
dren: Psychometric properties of a modified speech
sample. Br J Clin Psychol 42:53–67, 2003.
DeWolfe N, Byrne JM, Bawden HN: ADHD in preschool
children: Parent-rated psychosocial correlates. Dev
Med Child Neurol 42:825–830, 2000.
DuPaul GJ, McGoey KE, Eckert TL, VanBrakle J: Pre-
school children with attention-deficit/hyperactivity
disorder: Impairments in behavioral, social, and school
functioning. J Am Acad Child Adolesc Psychiatry 40:
Elliott CD: Differential Ability Scales. San Antonio
(Texas), Harcourt Assessment Corporation, 1990.
Eyberg SM, Boggs SR, & Algina J: Parent-child interac-
tion therapy: A psychosocial model for the treatment
of young children with conduct problem behavior and
their families. Psychopharmacol Bull 31:83–89, 1995.
Gadow KD, Sprafkin J: Child and Adolescent Symptom
Inventories. Odessa (Florida), Psychological Assess-
ment Resources, 1996.
Gordon M, Antshel K, Faraone S, Barkley R, Lewandowski
L, Hudziak JJ, Biederman J, Cunningham C: Symptoms
versus impairment: The case for respecting DSM-IV’s
Criterion D. J Attn Disorders 9:465–475, 2006.
Greenhill LL, Kollins SH, Abikoff H, McCracken J, Rid-
dle M, Swanson J, McGough J, Wigal S, Wigal T, Vi-
tiello B, Skrobala AM, Posner K, Ghuman J, Cunning-
ham C, Davies M., Chuang S, Cooper T: Efficacy and
safety of immediate-release methylphenidate treat-
ment for preschoolers with ADHD. J Am Acad Child
Adolesc Psychiatry 45:1284–1293, 2006.
Gresham FM, Elliott SN: Social Skills Rating System: Man-
ual. Circle Pines (Minnesota), American Guidance Ser-
Guy W: ECDEU Assessment Manual for Psychopharma-
cology: Revised. Washington (DC), U.S. Department of
Health, Education and Welfare (DHEW), 1976.
Hechtman L, Abikoff H, Klein RG, Greenfield B, Etcov-
itch J, Cousins L, Fleiss K, Weiss M, Pollack S: Chil-
FUNCTIONAL OUTCOMES IN THE PATS591
dren with ADHD treated with long-term methyl-
phenidate and multimodal psychosocial treatment: Im-
pact on parental practices. J Am Acad Child Adolesc
Psychiatry 43:830–838, 2004.
Kazdin AE, Whitley MK: Treatment of parental stress to
enhance therapeutic change among children referred
for aggressive and antisocial behavior. J Consult Clin
Psychol 71:504–515, 2003.
Kollins SH, Greenhill LL, Swanson J, Wigal S, Abikoff H,
McCracken JT, Riddle M, McGough JJ, Vitiello B, Wi-
gal T, Skrobala AM, Posner K, Ghuman JK, Davies M,
Cunningham C, Bauzo A: Rationale, design, and meth-
ods of the Preschool ADHD Treatment Study (PATS).
J Am Acad Child Adolesc Psychol 45:1275–1283, 2006.
Lahey BB, Pelham WE, Stein MA, Loney J, Trapani C, Nu-
gent K, Kipp H, Schmidt E, Lee S, Cale M, Gold E, Har-
tung CM, Willcutt E, Baumann B: Validity of DSM-IV
attention-deficit/hyperactivity disorder for younger
children. J Am Acad Child Adolesc Psychiatry 37:695–
McDermott PA, Steinberg CM, Angelo LE: Situational
specificity makes the difference in assessment of youth
behavior disorders. Psychol Sch 42:121–136, 2005.
Merrell KW, Wolfe TM: The relationship of teacher-rated
social skills deficits and ADHD characteristics among
kindergarten-age children. Psychol Sch 35:101–109,
Monteiro-Musten L, Firestone P, Pisterman S, Bennett S,
Mercer J: Effects of methylphenidate on preschool chil-
dren with ADHD: Cognitive and behavioral functions.
J Am Acad Child Adolesc Psychiatry 36:1407–1415,
MTA Cooperative Group: A 14-month randomized clini-
cal trial of treatment strategies for attention deficit/
hyperactivity disorder. Arch Gen Psychiatry 56:1073–
Murray DW, Kollins SH, Hardy KK, Abikoff HB, Swan-
son JM, Cunningham C, Vitiello B, Riddle MA, Davies
M, Greenhill LL, McCracken JT, McGough JJ, Posner
K, Skrobala AM, Wigal T, Wigal SB, Ghuman JK,
Chuang SZ: Parent versus Teacher Ratings of Atten-
tion-Deficit/Hyperactivity Disorder Symptoms in the
Preschoolers with Attention-Deficit/Hyperactivity
Disorder Treatment Study (PATS). J Child Adolesc
Psychopharmacol 17:605–619, 2007.
Schleifer M, Weiss G, Cohen NJ, Elman M, Cvejic H,
Kruger E: Hyperactivity in preschoolers and the effect
of methylphenidate. Am J Orthopsychiatry 45:38–50,
Shaffer D, Gould MS, Brasic J, Ambrosini P, Fisher P, Bird
H, Aluwahlia S: A Children`s Global Assessment Scale
(CGAS). Arch Gen Psychiatry 40:1228–1231, 1983.
Shaffer D, Fisher P, Dulcan MK, Davies M, Piacentini J,
Schwab-Stone ME, Lahey BB, Bourdon K, Jensen PS,
Bird HR, Canino G, Regier DA: The NIMH Diagnostic
Interview Schedule for Children Version 2.3
(DISC–2.3): Description, acceptability, prevalence
rates, and performance in the MECA study. J Am Acad
Child Adolesc Psychiatry 35:865–877, 1996.
Short EJ, Manos MJ, Findling RL, and Schubel EA: A
prospective study of stimulant response in preschool
children: Insights from ROC analyses. J Am Acad Child
Adolesc Psychiatry 43: 251–259, 2004.
Sonuga-Barke EJS, Daley D, Thompson M, Laver-Brad-
bury C, and Weeks B: A Parent-based therapies for pre-
school attention-deficit/hyperactivity disorder: A ran-
domized, controlled trial with a community sample. J
Am Acad Child Adolesc Psychiatry 40:402–408, 2001.
Sonuga-Barke EJS, Daley D, Thompson M, Swanson J: The
management of preschool AD/HD: Addressing un-
certainties about syndrome validity, diagnostic valid-
ity and utility, and treatment efficacy and safety. Ex-
pert Rev Neurother 3:465–476, 2003.
Sprafkin J, Volpe RJ, Gadow KD, Nolan EE, Kelly K : A
DSM-IV-referenced screening instrument for preschool
children: The Early Childhood Inventory-4. J Am Acad
Child Adolesc Psychiatry 41:604–612, 2002.
Stein MA, Blondis TA, Schnitzler ER, O’Brien T, Fishkin
J, Blackwell B, Szumowski E, Roizen NJ: Methyl-
phenidate dosing: Twice daily versus three times daily.
Pediatrics 98:748–56, 1996.
Swanson JM, McStephen M, Hay D, Levy F: The poten-
tial of the SWAN rating scale in the genetic analysis of
ADHD. Vancouver (BC), Presentation at the 10thMeet-
ing of the International Society for Research in Child
and Adolescent Psychiatry, 2001.
Vitiello B, Abikoff HB, Chuang SZ, Kollins SH, Mc-
Cracken JT, Riddle MA, Swanson JM, Wigal T, Mc-
Gough JJ, Ghuman JH, Wigal SB, Skrobala AM, Davies
M, Posner K, Cunningham C, Greenhill LL: Effective-
ness of methylphenidate in the 10-month continuation
phase of the Preschoolers with Attention-Deficit/
Hyperactivity Disorder Treatment Study (PATS).
J Child Adolesc Psychopharmacol 17:593–603, 2007.
Webster-Stratton C: Preventing conduct problems in
Head Start children: Strengthening parenting compe-
tencies. J Consult Clin Psychol 66:715–730, 1998.
Wigal T, Greenhill L, Chuang S, McGough J, Vitiello B,
Skrobala A, Swanson J, Wigal S, Abikoff H, Kollins S,
McCracken J, Riddle M, Posner K, Ghuman J, Davies
M, Thorp B, Stehli A: Safety and tolerability of
methylphenidate in preschool children with ADHD. J
Am Acad Child Adolesc Psychiatry 45:1294–1303, 2006.
Address reprint requests to:
Howard B. Abikoff, Ph.D.
NYU Child Study Center
215 Lexington Avenue, 14thfloor
New York, NY 10016
ABIKOFF ET AL.592
Reproducedwithpermissionofthecopyrightowner.Furtherreproductionprohibitedwithoutpermission. Download full-text