Side
Effects
of Methyiphenidate
in Children
With Attention
Deficit
Hyperactivity
Disorder:
A
Systemic,
Placebo-Controlled
Evaluation
Russell
A Barkley,
PhD;
Mary
B. McMurray,
MD;
Craig
S. Edelbrock,
PhD; and Kathryn
Robbins
From the Departments of Psychiatry, Pediatrics, and Neurology, University of
Massachusetts Medical Center, Worcester
ABSTRACT.
The frequency and severity of 17 side effectspresumably associated with stimulant medication were
assessed during a rigorous, triple-blind, placebo-con-trolled, crossover evaluation of methylphenidate, 0.3 and
0.5 mg/kg twice a day, in 83 children with attention
deficit hyperactivity disorder. Side effects were rated by
parents and teachers at the end of each weekly drug
condition. Three children (3.6%) had side effects that
were sufficiently serious to warrant immediate discontin-uation of medication. Parent ratings indicated that only
the side effects of decreased appetite, insomnia, stom-achaches, and headaches increased significantly in
fre-quency and severity during the two active medication
doses as compared with the placebo condition. Fewer
than half of the children experienced these side effects and among those who did, ratings of mean severity
re-mained in the mild range. Teacher ratings showed little
change over drug conditions, except on ratings of staring, sadness, and anxiety, which declined with increasing dose of medication. Surprisingly, a high frequency of these
behavior side effects were reported during the placebo
condition. Stimulant medication within this therapeutic
range, therefore, results in few, generally mild side effects.
However, the wide variation in individual responses and
the high rate ofthese behaviors during placebo treatment
argue for the systematic monitoring of side effects before
and during pediatric trials of stimulant medication.
Pe-diatrics 1990;86:184-192; attention deficit hyperactivity
disorder, methyphenidate, stimulant medication, side
ef-fects.
Stimulant medication, particularly methylphe-nidate (Ritalin), has become an increasingly com-mon treatment for the management of children
Received for publication Jun 16, 1989; accepted Nov 14, 1989. Reprint requests to (R.A.B.) Dept of Psychiatry, University of Massachusetts Medical Center, 55 Lake Ave N, Worcester, MA
01655.
PEDIATRICS (ISSN 0031 4005). Copyright © 1990 by the
American Academy of Pediatrics.
with attention deficit hyperactivity disorder (ADHD).’ Recent surveys2’3 estimate that between 3% and 6% of elementary-school-aged children are
taking medication for management of attention
deficits and hyperactivity, with stimulants repre-senting 99% of the medications prescribed. This represents an increase of up to 248% in the rate of prescribing such medication during the past 10 years.3 Along with this dramatic increase in medi-cation use has been a corresponding increase in the
extent of public interest in and controversy about
the medicating of children with ADHD for improv-ing their school performance and managing their behavior.4 Alarming stories have frequently ap-peared in the news media claiming that stimulant medication leads to murder, suicide, or substance abuse,” or that they are highly addictive drugs
often used as a “chemical billy club” or
“straitjacket” to turn normal, healthy children into “drug addicts,”12”3 Scientific meetings have been picketed as well, with some picketers even using elephants wearing signs or airplanes trailing
ban-ners stating “Psychs stop drugging our kids.”4
While many of these claims can be traced to a campaign by the Church of Scientology and its Citizens Commission on Human Rights5’6”5 to dis-credit the use of stimulant medication for school-children, they have made a substantial impact on the public’s attitude toward this medication. More-over, the Church and affiliated parents have threat-ened legal action against the American Psychiatric Association, claiming fraud in the diagnosis of
ADHD,’6”7 and they have threatened to sue as
many as 600 physicians for malpractice; at least 8
suits have already been filed as of May 1988.6
medica-tion, we surveyed the literature for figures on the
incidence of side effects among medicated children.
Although there exists a wealth of published
scien-tific research on the effects of stimulants on hyper-active children,’2#{176} we could locate no study that specifically and systematically evaluated the
behav-ioral side effects of medication. Most studies failed
to mention side effects at all as part of their results, or they related only anecdotal information about the most obvious side effects reported during the
drug trials. No studies have contrasted these rates
of behavioral side effects against those reported during the placebo conditions-a crucial
compari-son to establish the true incidence of these side
effects that can be directly attributed to the
medi-cation condition. We therefore undertook to study
systematically the incidence of side effects in a large
sample of children with ADHD undergoing a trial
of methylphenidate. These children were
partici-pating in a rigorous, triple-blind, placebo-controlled
evaluation of the effects of two therapeutic doses
(0.3 and 0.5 mg/kg twice a day) of methylphenidate
in the management of their ADHD symptoms.
A previous review and tabulation by Barkley’8 of the 29 studies through 1976 that casually reported the behavioral side effects of the stimulant
medi-cations found that insomnia was the most
fre-quently cited side effect (N = 26 studies), followed
in decreasing frequency by decreased appetite (N =
23), irritability (N = 13), weight loss (N = 12),
abdominal pains (N = 11), and headaches (N = 10).
Four studies each reported drowsiness and sadness as side effects, while 6 cases of stimulant-induced
psychosis were noted in the 110 studies reviewed,
which involved more than 4200 hyperactive
sub-jects. Yet, frequency of citation in the literature is
hardly a true indication ofthe actual incidence rates
of the various behavioral side effects among chil-then taking stimulant medication. For this reason, we were interested in tracking the occurrence of the
17 most common side effects attributed to
methyl-phenidate across both placebo and active drug
con-ditions during rigorous, blind, placebo-controlled evaluation of two therapeutic doses of methylphe-nidate.
METHODS
Subjects
A total of 83 children between 5 and 13 years of age served as subjects in this study. They had been
selected from children consecutively referred to an
ADHD clinic, part of the Departments of Psychia-try and Pediatrics at the University of
Massachu-setts Medical Center, and were considered
appro-priate candidates for a trial on stimulant
medica-tion. There were 12 girls and 71 boys. To be considered for the drug trial, a child had to meet the following criteria: (1) an estimated IQ score greater than 80 on the Peabody Picture Vocabulary Test21; (2) parental and/or teacher complaints of
significant problems with inattention, overactivity,
and impulsivity; (3) placement at the 93rd percent-ile on the hyperactivity scales of the parent or teacher version of the Child Behavior Checklist22; (4) appearance of symptoms before age 7 years; (5)
symptoms persisting for at least 12 months; and (6)
no gross sensory or motor disabilities, epilepsy,
autism, psychosis, tic disorders or Tourette
syn-drome, or significant cardiac problems. All parents of subjects signed written informed consent state-ments agreeing to participate in this drug protocol. The subject characteristics are displayed in Table 1.
Procedures
All children received an initial medical and psy-chologic evaluation using a variety of tructured interviews, rating scales, and laboratory measures of ADHD symptoms, the results of which are re-ported elsewhere.23 These children then partici-pated in a triple-blind, placebo-controlled, cross-over evaluation in which each subject was tested on placebo, low-dose (0.3 mg/kg) methylphenidate, and high-dose (0.5 mg/kg) methylphenidate given twice daily at morning and noon. Each medication condition lasted 7 to 10 days before crossing over
TABLE 1. Children With Attention Deficit
Hyperac-tivity Disorder: Initial Selection Measures*
Measure Mean (SD)
Age,y 8.2 (2.2)
Education, y 3.6 (2.2)
IQ (PPVT) 105.1 (15.3)
Mother’s age, y 34.6 (5.8)
Mother’s education, y 13.6 (2.1)
CBCL scales
Aggression 70.5 (6.6)
Hyperactivity 75.3 (8.8)
CPRS scales
Conduct Problems 13.8 (6.3)
Impulsive-Hyperactive 8.1 (3.2)
Hyperactive Index 17.8 (5.2)
CBCL-TRF scales
Inattention
68.5
(7.0)Nervous-Overactive
68.5
(9.0)CTRS scales
Conduct Problems 8.3 (5.6)
Hyperactivity 12.1 (5.6)
Inattentive 14.1 (4.8)
Hyperactive Index 17.4 (8.3)
* PPVT, Peabody Picture Vocabulary Test; CBCL, Child
Behavior Checklist; CPRS, Conners Parent Rating Scale; CBCL-TRF, Child Behavior Checklist-Teacher Report
TABLE 2.
Mean Severity Ratings by Parents for Each of 17 Side Effects for Each DrugCondition*
Placebo Low Dose High Dose F P Contrast
0.4 1.7 2.6 25.9 <.01 abc
1.4 2.6 3.1 13.4 <.01 ab
0.5 1.0 1.5 9.4 <.01 abc
0.3 0.6 0.8 4.4 <.05 ab
1.7 2.2 2.0 0.9 . . . ...
0.7 0.9 1.1 1.6 . . . ...
* F, value of the F test from the analyses of variance with repeated measures; P, level of
statistical significance of the overall F test; contrast, the statistically significant pairwise contrasts between the drug conditions on those measures having a significant overall main
effect for drug condition where a = placebo vs low dose, b = placebo vs high dose, and c
= low dose vs high dose.
to the next condition. Children were randomly as-signed to one of six possible orders of drug condi-tions, with similar numbers of subjects participat-ing in each drug order. The medication was pre-pared by the hospital pharmacy to the nearest 2.5-mg dose based on the children’s body weight and placed within orange, opaque gelatin capsules to
disguise the distinctive differences in flavor
be-tween medication and placebo and the dose differ-ences across conditions. Unused capsules were re-turned to the clinic for a check on adherence to the medication schedules each week. No family was discontinued from the study because of noncompli-ance, defined as more than 1 day of failure to use the medication. The children, their parents and teachers, and the research assistant evaluating the children each week were blind to the medication order.
At the end of each week, the children were tested for 45 minutes on a brief battery of laboratory measures of attention, impulsivity, verbal learning and memory, and ADHD symptoms during aca-demic performance. Their parents and teachers also completed a packet of behavior-rating scales each week. These results are reported elsewhere.23 As part of this packet, parents and teachers completed a questionnaire on the side effects of methylphe-nidate (see Appendix). The questionnaire lists the
17 most common side effects of methylphenidate
and asks parents to rate their occurrence each week on a scale from 0 (not a problem) to 9 (severe). The rating scale was labeled generically as a “Behavior
Questionnaire” to disguise its intended use as a
monitoring tool for potential side effects. The pur-pose was to prevent prejudice on the part of the respondents, who might potentially distort their
ratings if they knew these could be side effects of
the medication.
RESULTS
Three children (3.6% of sample) were unable to complete the drug protocol because of serious ad-verse reactions to medication. One child had a nervous facial tic, dizziness, and headache; a second had dizziness, headache, and increased hyperactiv-ity; and the third had excessive speech and dis-jointed thinking. The magnitude of these reactions was sufficient to concern both the parents and clinicians and medication was immediately discon-tinued.
The mean severity ratings for these parent-rated side effects across each drug condition were sub-mitted to one-way analyses of variance with re-peated measures. The means, standard deviations, and statistical test results are shown in Table 2. The results indicated that of the 17 side effects, only 4 showed significant drug effects, and these were insomnia, decreased appetite, stomachaches, and headaches. Subsequent pairwise comparisons of the drug conditions for each of these side effects indicated that both the low dose and high dose resulted in significant increases in these four side effects relative to the placebo. However, only on
Side Effect
Decreased appetite
Insomnia
Stomachaches Headaches
Prone to crying Tics/nervous move-ments
Dizziness Drowsiness Nail biting Talks less
Anxious
Disinterested in others Euphoria
Irritable
Nightmares Sadness Staring
0.1 0.3 0.3 1.6
0.5 0.7 0.7 0.5
1.1 1.1 1.3 0.8
0.4 0.6 0.9 1.9
2.4 1.8 2.0 1.7
0.7 0.6 0.6 0.2
1.7 1.4 1.8 0.7
3.2 2.5 2.7 1.6
0.6 0.6 0.8 1.4
1.6 1.9 1.8 0.3
15 52 56
1 7 13
40 62 68
7 18 18
18 39 35
0 1 6
11 26 21
0 1 4
49 59 54
10 16 10
18 18 28
4 7 5
4 10 7
0 0 1
18 23 20
1 2 1
22 26 29
7 4 9
16 20 22
1 1 2
58 58 52
12 9 7
18 18 15
0 1 2
41 34 43
9 4 7
72 65 66
18 15 13
20 20 21
0 0 3
43 48 41
5 6 8
ratings of decreased appetite and stomachaches did
the high dose produce a significantly greater in-crease as compared with the low dose. For ratings of insomnia and headaches, the two doses were not
significantly different from each other.
Another method of examining the behavioral side
effects of medication is to calculate the percent
occurrence of the side effects in each drug condition to yield clinically useful information about the in-cidence rates one can expect in practice. We there-fore calculated the percentage of children in each
drug condition who had ratings of 1 or higher on
each side effect as a measure of percent occurrence. We also computed the percentage who had parent ratings of 7 or higher as a measure of how many
subjects had serious or severe occurrences of these
side effects. These percentages across the placebo
and medication doses are shown in Table 3. In this analysis, it can be seen again that the most frequent side effects on parent ratings are
decreased appetite, insomnia, stomachaches, and
headaches, in descending order of percent change
over placebo rates. However, the percentage of chil-then experiencing proneness to crying also
in-creased by at least 10% during the low-dose
condi-tion, while the percentage reporting tics/nervous movements increased by 10% at the high dose of medication (both are statistically significant, P <
.05). For the remaining 10 side effects, no signifi-cant medication effects were noted for either dose
condition. The percentage of children with ratings
in the serious range (7 or higher) increased a small but significant degree on the ratings of decreased appetite and insomnia, such that at the high dose,
13% and 18% of the sample, respectively, had
se-rious reactions, as compared with only 1% and 7%,
respectively, for these side effects during the pla-cebo condition.
Complete data from teacher ratings of side effects
across all three drug conditions were available for
a total of 53 subjects. Subjects for whom such data were not available were either tested during the
summer months or school vacations or had
incom-plete school data for all three drug conditions. The
severity ratings for each side effect were submitted
to a one-way analysis of variance with repeated
measures. The mean severity ratings across the
three drug conditions as well as the results of these
statistical tests are shown in Table 4, along with
the results of pairwise contrast analyses conducted on those measures having a significant main effect for drug condition. These results indicated that medication effects were found for only 3 of the 17
behavioral side effects, these being
staring/day-dreaming, sadness and anxiety. In each case, the
medication resulted in significant declines in these
TABLE 3.
Percentage of 82 Subjects Displaying Eachof 17 Side Effects of Methylphenidate During Each Drug
Condition (Parent Ratings)*
Side Effect Placebo Low Dose High Dose (0.3 mg/ (0.5 mgI
kg) kg) Decreased appetite % % severe Insomnia % % severe Stomachaches % % severe Headaches % % severe
Prone to crying % % severe Tics/nervous movements % % severe Dizziness % % severe Drowsiness % % severe Nail biting % % severe Talks less % % severe Anxiousness % % severe
Disinterested in others
% % severe Euphoria % % severe Irritable % % severe Nightmares % % severe Sadness % % severe Staring % severe
40 38 38
2 4 1
* % refers to the percentage of subjects in whom the side
effect was rated 1 or higher on the scale of severity (1 to
9); % severe refers to the percentage of subjects in whom the side effect was rated 7 or higher.
severity ratings. For staring, both the low and high
doses resulted in significant declines compared with
placebo but did not differ from each other. For
TABLE 4.
Mean Severity Ratings by Teachers for Each of 17 Side Effects for Each DrugCondition*
Placebo Low Dose High Dose F P Contrast
0.6 0.4 0.9 1.3 . . . ...
0.2 0.1 0.3 0.3 .. . ...
0.5 0.5 0.7 0.9 . . . ...
0.4 0.3 0.5 0.2 . . . ...
0.7 0.5 0.2 2.4 . . . ...
1.6 1.1 1.0 2.0 . . . ...
<.05
<.01
b
be
ab
* F, value of the F test from the analyses of variance with repeated measures; P, level of
statistical significance of the overall F test; contrast, the statistically significant pairwise contrasts between the drug conditions on those measures having a significant overall main
effect for drug condition where a = placebo vs low dose, b = placebo vs high dose, and c
= low dose vs high dose.
Side Effect
Decreased appetite Insomnia
Stomachaches
Headaches
Prone to crying Tics/nervous
move-ments
Dizziness Drowsiness Nail biting
Talks less
Anxious
Disinterested in others
Euphoria
Irritable Nightmares Sadness
Staring
0.1 0.1 0.0 0.6
1.0 0.9 0.9 0.1
0.6 0.3 0.4 1.2
1.5 1.4 1.4 0.0
2.4 1.9 1.5 3.4
1.3 1.3 1.4 0.0
2.2 1.8 1.6 1.2
1.5 1.3 1.0 1.5
0.0 0.0 0.1 1.0
1.7 1.6 1.0 3.1
3.0 2.1 2.0 4.1
declines relative to both the placebo and low-dose conditions. For ratings of anxiety, only the high dose resulted in declines compared with the placebo condition.
We calculated the percentage ofthese 53 children who showed any occurrence of each behavioral side effect as well as the percentage having ratings in the severe range (7 or higher). These percentages
are shown in Table 5. Declines in the percentage of subjects having the side effect were noted on teacher ratings of staring/daydreaming, irritability, sadness, proneness to crying, and anxiety. Similar trends for decreasing incidence of side effects with medication were noted for the percentage of sub-jects with ratings in the severe range.
DISCUSSION
In this paper, we have attempted to provide a more systematic analysis of the side effects occur-ring in children with ADHD treated with therapeu-tic doses of methyiphenidate. Our results are con-sistent with the earlier tabulation of studies by Barkley’8 in showing that decreased appetite and insomnia are the most common side effects of stim-ulant medication reported by parents, occurring to a significantly greater degree and in a greater per-centage of children than in the placebo condition. Whereas only 15% of our subjects apparently ex-perienced insomnia as troublesome during the pla-cebo condition, this increased to 52% and 56% in the low- and high-dose conditions, respectively-a
change of37% to 41% as compared with the placebo percentages. The percentage of children with in-somnia rated in the serious range also increased substantially, from only 1% during the placebo condition to 13% during the high-dose condition. Appetite suppression increased to a less striking
degree, from 40% during the placebo condition to
62% and 68% in the low- and high-dose conditions, respectively. Here again, the percentage of subjects with ratings in the serious range for this side effect increased from 7% to 18% between placebo and both doses of medication. Irritability, however, was the next most commonly reported behavioral side effect in that tabulation but was not found in our study to occur significantly more during drug than in placebo conditions. Consistent with earlier re-views of the literature, we found abdominal aches and headaches to be the next most common
behav-ioral side effects of methylphenidate. Stomachaches
increased from 18% to 39% over the placebo
con-dition during both the low- and high-dose condi-tions, more than doubling in percent occurrence during medication treatment. Headaches also more than doubled in frequency of occurrence, from 11% during the placebo condition to 21% to 26% during active medication treatment.
TABLE 5. Percentage of 53 Subjects Displaying Each
of 17 Side Effects of Methylphenidate During Each Drug
Condition (Teacher Ratings)*
Side Effect Placebo Low Dose High Dose
(0.3 mg/ (0.5 mgI
kg) kg)
17 15 21
2 2 6
4 6 6
2 0 2
17 17 17
0 2 2
Decreased appetite %
% severe Insomnia
% % severe Stomachaches
% % severe Headaches
% % severe Prone to crying
% % severe
Tics/nervous movements %
% severe Dizziness
% % severe Drowsiness
% % severe Nail biting
% % severe Talks less
% % severe Anxiousness
% % severe
Disinterested in others %
% severe Euphoria
% % severe Irritable
% % severe Nightmares
% severe Sadness
% % severe Staring
% % severe
* % refers to the percentage of subjects in whom the side effect was rated 1 or higher on the scale of severity (1 to 9); % severe refers to the percentage of subjects in whom the side effect was rated 7 or higher.
conditions. Moreover, teacher ratings of these side effects-especially those behaviors claimed by
crit-ics to be common negative reactions, such as
star-ing, disinterest in others, sadness, and
anxiety-actually decreased during drug treatment. In gen-eral, this study corroborates past conclusions based on more anecdotal, clinical reports that the stimu-lant medications, when given in therapeutic doses, are generally safe and produce only minor degrees of side effects in most children with ADHD. The
results lend no support to the recent wave of media
stories that stimulant medication produces serious
behavioral or emotional disturbance as a common consequence in the treatment of children with
ADHD. Only three of the 83 children had side
effects that were sufficiently serious to warrant immediate cessation of medication, and in all three
cases the side effects disappeared within the day.
1 i; Our results are surprising in several respects.
First, they suggest that many of the behavioral side
21 11 4 effects attributed to stimulant medication already
0 4 0 occur in a sizeable percent of the population of
30 25 26 children with ADHD before they begin drug
ther-11 8 4 apy. In this sample, parent ratings showed that
40% had problems with insomnia, 49% with
prone-6 4 2 ness to crying, 72% with irritability, 58% with
0 0 0 anxiousness, 43% with sadness or dysphoric mood,
23 21 25 and 40% with staring during the placebo condition.
4 4 2 Similarly high rates were seen in teacher ratings
during this placebo condition for these behaviors.
13 13 8 These percentages of occurrence cannot be easily
2 2 4 attributed to carryover effects from the medication
38 42 38 conditions as such effects dissipate within the 24
4 2 7 hours after discontinuation of the active
medica-tion. It would therefore seem that at least some of
55 49 42 the previously reported instances ofthese particular
9 6 6 “side effects” during stimulant medication
treat-36 43 30 ment may have been confounded with high
base-4 0 9 rate occurrences of these behaviors in this
popula-tion. At the very least, these results argue for
sys-45 42 40 tematic monitoring of side effects both before and
8 8 6 during medication treatment to discriminate more
45 34 25 clearly a true drug side effect from a preexisting 9 7 2 condition. Clinicians are encouraged to use a rating
scale similar to ours as part oftheir clinical protocol
0 0 2 for evaluating stimulant drug responding in
chil-0 0 0 dren with ADHD and to incorporate placebo
con-40 38 30 trols where possible as a further means of
evaluat-9 11 2 ing actual medication effects.
A second unexpected result was the relatively
68 57 57 high percentage of children whose parents reported
19 11 9 tics or nervous movements during both the placebo
and active medication conditions. At least 18% of our sample apparently had such problems during the placebo condition, which is somewhat higher than the acknowledged 12% occurrence of simple
tics in childhood reported in other studies.24 Our
of this rating scale, it would seem prudent to sepa-rate these two behaviors to delineate more clearly the specific side effects of the medication on mdi-vidual patients. The increase of 10% between the placebo and high-dose conditions is also somewhat higher than previous reports of medication-induced tics, which have ranged from 1% to 5%,25 but it replicates our earlier finding6 that medication-in-duced tics occurred in 10% of a previous sample of more than 30 children with ADHD undergoing a
similar evaluation protocol. Again, our figures could
be somewhat higher because of the confounding of
“tics” with “nervous movements” in the same item
on the questionnaire. These findings suggest that
closer monitoring of this behavioral side effect
should occur during clinical trials with children but
that there is no cause for alarm as all instances of
tics or mannerisms noted during medication
con-ditions subsided once medication ceased. It would seem wise to follow the recommendations of others
that stimulant medication be used cautiously or not
at all in children with a history of nervous tics or Tourette syndrome. Furthermore, when tics de-velop during a medication trial, the medication
should be discontinued, which generally
amelio-rates the problem.’8’27
A third surprising finding was the significant decline in several side effects based on teacher ratings during school. Staring or daydreaming, sad-ness, and anxiety all decreased, particularly during the high-dose condition, indicating that
methylphe-nidate produces positive improvements in these
domains of academically disruptive behavior rather
than serving as a “chemical billy club” to turn children into “zombies.”2”3 However, it must be recognized that teachers are not especially sensitive
reporters of mild affective symptoms in children and different results might have been obtained had the children been interviewed. Some studies have reported observations of mild dysphoria when trained behavior coders have reviewed videotapes of children with ADHD, with and without medica-tion, although other studies have not.3#{176}Many children dislike taking medication,3’ perhaps be-cause of these mildly unpleasant feelings. It is clear, however, that such dysphoric effects occur incon-sistently and are mild when they do occur.
In contrast with teacher ratings of a decline in
several side effects, parents reported that several
side effects actually increased in the home
environ-ment. These contrasting results suggest that the
side effects of stimulant medication occur primarily during the washout phase of the time course of the medication, when parents rather than teachers are
likely to be caring for the children. Even so, the
side effects seen during this rebound period are
often mild and widely variable across children. As
a consequence of these discrepancies in the situa-tional occurrence of side effects, parents and
teach-ers may sharply disagree over the value of the
medication in treating ADHD symptoms, where teachers see the primary behavioral benefits while parents see the side effects.32
Although our results, on the whole, underscore the relative safety of methylphenidate when used within the therapeutic ranges employed here, the wide individual variability in drug response
re-ported by others33’34 was typical of this large sample as well. A few children experienced more serious
increases in side effects during active medication
than these group means suggest, with 19% showing minimal or poor responding after completing the entire protocol and 3.6% discontinuing their partic-ipation before completing the drug trials because of
the side effects.23 The percent of children showing minimal or poor responding is similar to that found in our previous study26 of a separate, smaller group
(N = 23) of children with ADHD, while the number who had to terminate the protocol prematurely is half that seen in our earlier study (7.1%). It is not clear why fewer children had to discontinue medi-cation prematurely in the present study. Some
sub-jects in the present study, despite positive
improve-ments in behavior while taking our higher dose, had an accompanying increase in side effects noted at this dose level that resulted in our recommending it for fewer than 25% of this sample, with a lower dose being suggested for the remainder.
The relative safety of the medication in our group should, therefore, not obscure the need to monitor closely all children taking such medication because of this unpredictability in adverse responding that may occur in some individual cases. Again, the inclusion of a simple rating scale of side effects,
such as that used here, along with a brief set of
parent and teacher rating scales of ADHD
symp-toms as recommended by others,26’’36 would greatly
enhance the ability of the pediatrician to detect and monitor both desirable and undesirable behavioral
changes resulting from stimulant medication trials.
It would also reassure parents, in this day of “anti-Ritalin” sentiment in the media, that proper mon-itoring was being done, and it would provide crucial
integration of information from school, not usually
obtained in medication trials,32 with that from home and office. In our experience, teachers are better evaluators ofpositive drug responses because
of their observations of children during the active
time course of the medication, while parents are
better at assessing side effects during the
late-afternoon “washout” or rebound phase of these
medications.26
be almost impossible to obtain in clinical practice. Finally, it must be noted that other side effects may have occurred that simply were not listed on this rating scale and went undetected. However, when combined with the Conners rating scales of child behavior, the side effects rating scale provides a reasonably comprehensive picture of children’s
be-havioral adjustment. Had serious side effects
oc-curred in areas not rated on these scales, we would have expected to be informed of them during the weekly office visits during the drug trial.
Our findings are encouraging in that where proper screening, diagnosis, and monitoring of chil-dren with ADHD are incorporated in clinical trials of stimulant medication, rates of side effects will be generally low and those that occur will be relatively mild. Where more serious side effects occur, de-creasing or discontinuing the medication amelio-rates the problem. Clinicians can continue the stim-ulant drug treatment of appropriate children with
ADHD knowing that many recent reports in the media of adverse Ritalin responses are rare and unusual cases that do not reflect the actual rates of occurrence of the typical side effects of these drugs.
APPENDIX:
STIMULANT
DRUG SIDE EFFECTS
RAT-ING SCALE
Name _____________________ Date ______________
Person Completing This Form __________________
Instructions
Please rate each behavior from 0 (absent) to 9 (serious).
Circle only one number beside each item. A zero means that you have not seen the behavior in this child during the past week, and a 9 means that you have noticed it
and believe it to be either very serious or to occur very frequently.
Behavior Absent Serious
Insomnia or trouble 0 1 7 8 9
sleeping
Nightmares 0 1
Stares a lot or day- 0 1
dreams
Talks less with others 0 1
Uninterested in oth- 0 1
study are based on a large sample that was screened
before admission to the drug trial. Had all referred children been submitted to methylphenidate test-ing, it is likely that the rate of side effects and premature discontinuation during the trials would have been somewhat higher. As recommended by many authorities in this field,’9’20’27 we routinely use a multimethod assessment of children; we use mul-tiple sources of information and screen out children with the following conditions because of their known association with adverse responding: (1) tics, Tourette syndrome, or a immediate family history of such; (2) thought disturbance or schizoid, schizotypal, or frank psychotic features; (3) cardiac
disorders or high blood pressure; (4) excessive levels
of anxiety or fears; (5) levels of depression that equal or exceed the problems with ADHD; and (6) clear-cut evidence of brain damage. This is not to
say that such children may not, in extreme cases,
warrant a medication trial but to caution those who would undertake such a trial to use far lower initial
doses than those routinely used in children with
ADHD and to proceed with much finer dose
incre-ments and behavioral monitoring during clinical
titration. There is no evidence that children with
epilepsy have their seizure disorder adversely
af-fected by stimulant medication, so where ADHD coexists with epilepsy, a medication trial can be considered as relatively safe.37 However, we are aware of as yet unpublished data (Fischer M. 1989) on a large sample of epileptic children with ADHD in which the rates of behavioral side effects were considerably higher than those in matched, none-pileptic children with ADHD, again suggesting closer monitoring of these children during medica-tion trials.
Another limitation was the range of medication chosen for study. It is likely that had higher doses of medication been used, especially above 0.7 mg/ kg, greater side effects would have been noted.33’34’38 However, our doses represent those typically used in clinical practice, so these rates of side effects are a reasonable estimation of what can be expected in clinical trials that use doses within the therapeutic
range.
A third limitation is the use of a rating scale to monitor side effects rather than direct behavioral observations. Rating scales are prone to response bias from the respondent, being but a quantified opinion of a caretaker in the child’s environment.39 For this and other reasons, we chose not to actually label the rating scale as a measure of “side effects” but simply as a “behavior questionnaire.” Despite
such problems, rating scales do provide an
inexpen-sive means of collapsing observations of the child’s
drug response by those who best know the child
across lengthy time periods that would otherwise
ers
Decreased appetite
Irritable
Stomachaches Headaches
Drowsiness Sad/unhappy Prone to crying
Anxious Bites fingernails
Euphoric/unusually happy
Dizziness
Tics or nervous move-ments
23456
23456789
23456789
23456789 23456789
0123456789 0123456789 0123456789 0123456789 0123456789 0123456789 0123456789 0123456789 0123456789 0123456789
REFERENCES
1. American Psychiatric Association, Committee on Nomen-clature and Statistics. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed, revised. Washington, DC: Ameri-can Psychiatric Association; 1987
2. Safer DJ, Krager JM. A survey of medication treatment for hyperactive/inattentive students. JAMA 1988;260:2256-2258
3. Safer DJ, Krager JM. Trends in medication treatment of hyperactive school children. Clin Pediatr (Phila). 1983;22:500-504
4. Toufexis A. Worries about overactive kids: are too many youngsters being misdiagnosed and medicated? Time. Jan-uary 16, 1989:65
5. Bass A. Debate over Ritalin is heating up: experts say critics are lashing out for all the wrong reasons. Boston GlObe.
March 28, 1988:36-38
6. Cowart VS. The Ritalin controversy: what’s made this drug’s opponents hyperactive? JAMA. 1988;259:2521-2523
7. Bacon J. What’s the best medicine for hyper kids? USA Today. February 17, 1988:4D
8. Williams L. Parents and doctors fear growing misuse of drug used to treat hyperactive kids. Wall Street Journal. January 15, 1988:18
9. Ritalin linked to bludgeoning death of teenager. The Coil.
March 8, 1988:3
10. Laccetti S. Parents who blame son’s suicide on Ritalin use will join protest. Atlanta Journal. August 13, 1988:B1,B7 11. Rogahn K. Ritalin: is it overprescribed? Those who use it
say life is hard without it. The Gazette. December 4, 1988:1A,13A
12. Dockx P. Are school children getting unnecessary drugs?
Sun Chronicle. January 11, 1988:15
13. McCartney S. Success may be encouraging excess of behav-ior medication drug Ritalin. Main Weekend. April 9, 1988:Business-2
14. Barkley R. Ritalin, Russia, and other ruminations. Newslet-ter of Section on Clinical Child PsycholOgy, Division 12, American PsycholOgical Association. 1988;3:1
15. Citizens Commission on Human Rights. Ritalin: A Warning for Parents. Los Angeles, CA: Church of Scientology; 1987 16. Henig RM. Courts enter the hyperactivity fray: the drug
Ritalin helps control behavior, but is it prescribed need-lessly? Washington Post Health. March 15, 1988:8
17. Twyman AS. Use of drug prompts suit. Newton Graphic.
May 4, 1988:28
18. Barkley RA. A review of stimulant drug research with hy-peractive children. J Child Psychol Psychiatry 1977;18:137-165
19. Gadow K. Children on Medication. Greenwich, CT: JAI Press;1983;1,2
20. Donnelly M, Rapoport JL. Attention deficit disorders. In: Weiner JM, ed. Diagnosis and PsychopharmacolOgy of Child-hood and Adolescent Disorders. New York: Wiley;
1985:178-198
21. Dunn L, Dunn L. The Peabody Picture Vocabularj Test.
Circle Pines, MN: American Guidance Service; 1981 22. Achenbach T, Edelbrock CS. Manualfor the Child Behavior
Checklist. Burlington, VT: Achenbach T, Edelbrock CS;
1983
23. Barkley RA, McMurray MB, Edeibrock CS, Robbins K. The effects of two doses of methylphenidate on aggressive and non-aggressive ADHD children. J Am Acad Child Adolesc Psychiatry. 1989;28:873-881
24. Barkley RA. Tic disorders and Tourette’s syndrome. In: Mash EJ, Terdal L, eds. BehavioralAssessment of Childhood Disorders. New York, NY: Guilford Press; 1988:552-585 25. Denckla MB, Bemporad JR, MacKay MC. Tics following
methylphenidate administration: a report of 20 cases. JAMA. 1976;235:1349-1351
26. Barkley HA, Fischer M, Newby R, Breen M. Development of a multimethod clinical protocol for assessing stimulant drug response in children with attention deficit disorder. J Clin Child Psychol. 1988;17:14-24
27. Golden G. The effect of central nervous system stimulants on Tourette’s syndrome. Ann NeuroL 1977;2:69-70
28. Cantwell D, Carlson G. Stimulants. In: Werry J, ed. Pedi-atric PsychopharmacolOgy. New York, NY: Brunner/Mazel; 1978:171-207
29. Whalen CK, Henker B, Granger DA. Ratings of medication effects in hyperactive children: viable or vulnerable? Behav Assess. 1989;11:179-200
30. Whalen CK, Henker B, Buhrmester D, Hinshaw SP, Huber
A, Laski K. Does stimulant medication improve the peer
status of hyperactive children? J Consult Clin PsychoL
1989;57:545-549
31. Baxley GB, Turner P. Hyperactive children’s knowledge and attitudes concerning drug treatment. J Pediatr Psychol.
1978;3:172-176
32. Gadow KD. Pharmacotherapy for behavior disorders. Clin Pediatr (Phila). 1983;22:48-53
33. Rapport MD, DuPaul GJ, Stoner G, Jones JT. Comparing classroom and clinic measures of attention deficit disorder: differential, idiosyncratic, and dose-response effects of methylphenidate. J Consult Clin PsychoL 1986;54:334-341 34. Douglas VI, Barr RG, Amin K, O’Neill ME, Britton BG.
Dosage effects and individual responsivity to methylpheni-date in attention deficit disorder. J Child PsycholPsychiatry. 1988;29:453-476
35. Ottinger DR, Halpin B, Miller M, Demian L, Hannemann R. Evaluating drug effectiveness in an office setting for children with attention deficit disorders. Clin Pediatr (Phiki). 1985;24:245-251
36. Ullmann RK, Sleator EK. Responders, nonresponders, and
placebo responders among children with attention deficit disorder. Clin Pediatr (Phikz). 1986;25:594-599
37. Crumrine PK, Feldman HM, Teodori J, Handen BL, Alvin RM. The use of methylphenidate in children with seizures and attention deficit disorder. Ann NeuroL 1987;22:441-442 38. Sprague RL, Sleator EK. Drugs and dosages: implications
for learning disabilities. In: Knights RM, Bakker DJ, eds.
The Neuropsychology ofLearning Disorders. Baltimore, MD: University Park Press; 1976:351-366
