In this issue...

Classroom behavior of children with ADHD: Relationship to
gender and the presence of other difficulties

New information on ADHD in girls

---

As one would expect, significant differences between the classroom behavior of children with ADHD and without ADHD have been reported in a number of studies.  These studies, however, have typically included few if any females.  As a result, objective information about classroom behavior differences between boys and girls with ADHD is quite limited.  This is an important gap in the literature, as better descriptive information on behavior differences between boys and girls with ADHD could result in more accurate identification of ADHD in girls, who often are identified later than boys or missed altogether.

Information on this under-researched topic was presented recently in a paper based on the MTA study (Abikoff, et al., (2002). Observed classroom behavior of children with ADHD: Relationship to gender and comorbidity. Journal of Abnormal Child Psychology, 30, 349-359). (Note: the MTA study is the largest treatment study of ADHD ever conducted.  You can read about previously published results from this study by going here and here.)

Participants in this observational study were 502 7-10 year old children diagnosed with ADHD, Combined Type (403 boys and 99 girls) recruited at all 6 study sites.  The results reported below thus reflect observations that occurred in many classrooms in multiple regions of the country.

Children were observed on two separate occasions in their classroom by trained observers.  Each observation period lasted 16 minutes.  (Note:  For approximately 10% of children, only a single 16-minute observation was conducted).  For each child with ADHD observed, a classmate of the same gender and ethnicity as the ADHD child was observed.  This comparison child was identified by the teacher as a "typical classmate" - i.e. no excessive behavior problems but not the best-behaved student either.

Observations occurred during teacher-led lessons and/or independent academic seatwork time under teacher supervision and observers were informed about specific classroom rules prior to the observation.  All children had been diagnosed relatively recently, and observations were conducted prior to the start of either medical and/or behavioral treatment provided in the study.  To keep observations from being biased, observers did not know which children had been diagnosed with ADHD and which were comparison children.

Each ADHD student and his/her comparison child were observed in 4 alternating 4-minute blocks, yielding 16 minutes of data on each child.  During this time, observers were trained to note the occurrence of the following behaviors: interference (e.g., clowning, interrupting others, talking during work), interference to teacher (e.g. interrupting teacher), off-task (sustained inattention or distractibility), noncompliance (not complying with teacher requests or instructions), aggression (physical aggression or destruction of property), verbal aggression to children (e.g., name calling, taunting, teasing), verbal aggression to teacher (e.g. name calling, arguing), minor motor movement (e.g. rocking in seat), gross motor - vigorous (e.g., running, skipping), out-of-chair (extended time out of seat), and solicitation (e.g. going up to teacher, calling out to teacher).

In addition to these individual behavior codes, two composite scores were calculated: the two gross motor categories were combined to yield a composite gross motor score, and, an overall ADHD composite was obtained by summing scores for interference, interference to teacher, off task, and the gross motor composite.

As noted above, nearly all children were observed twice, resulting in 32 minutes of data for over 90% of the sample.  Observation periods were divided into 15-second intervals, and the initial occurrence of each behavior observed during the interval was recorded.  Thus, multiple behaviors could be recorded for a single interval, but each behavior would be counted only once per interval.  Off-task, non-compliance and out-of-chair behaviors, had to be observed for the entire 15-second interval to be recorded.  When none of the problem behaviors occurred during an interval, the code "absence of behavior" was recorded.

Summary scores were derived for each behavior by calculating the percentage of intervals in which it was observed.  For example, 32 minutes of observation translates into 128 15-second intervals.  If a child were off-task during 22 of those intervals, his off-task score would be 22 divided by 128 or 17%.

RESULTS
 

BOYS VERSUS GIRLS

The researchers first compared behaviors for boys and girls overall, irrespective of ADHD status.  Boys had significantly higher rates of observed interference, and also had higher scores on the gross motor composite and ADHD composite.  Differences on other behaviors were consistently higher for boys, but were not statistically significant.
 

ADHD BOYS VERSUS COMPARISON BOYS

ADHD boys had higher observed rates of all behaviors coded than comparison boys except for solicitation of the teacher.  To put these differences in context, behaviors comprising the ADHD composite (i.e. interference, interference to teacher, off task, gross motor - standing, and gross motor - vigorous) were observed in 38% of the intervals for ADHD boys versus 15% of the intervals for comparison boys.  Non-compliance and aggressive behavior was observed to occur infrequently in both groups, but was still more common for ADHD boys than comparison boys (i.e. 2.4% vs. .3% for non-compliance; .5% vs. .1% for physical aggression; .9% vs. .3% for verbal aggression).  One reason why frequency of aggressive behavior may have been so low is that all observations occurred in structured classroom settings.  In less supervised settings (e.g. lunch room, playground) the occurrence of aggression would likely have been greater.
 

ADHD GIRLS VERSUS COMPARISON GIRLS

Girls with ADHD differed with comparison girls on all but three of the behavior categories (i.e. physical aggression, verbal aggression to teacher, and out-of-seat behavior).  Behaviors comprising the ADHD composite were observed in 25% of the intervals for ADHD girls versus 11% of intervals for comparison girls.
 

ADHD BOYS VERSUS ADHD GIRLS

Compared to girls with ADHD, boys with ADHD had significantly higher rates of interference (21.4% of intervals vs. 12.5), total aggression (1.9% vs. .9%), gross motor composite (3.5% vs. 2.3%), and the ADHD composite (38% vs. 25%).
 

EFFECT OF COMORBID PROBLEMS

In addition to examining behavior differences in relation to ADHD status and gender, the authors also examined whether comorbid behavior disorders (i.e. an additional diagnosis of either oppositional defiant disorder (ODD) or conduct disorder (CD) or an anxiety disorder related to classroom behavior.  Of the entire sample of children with ADHD, approximately 12% had ADHD + anxiety, 34% had ADHD + ODD/CD, 22% had ADHD + anxiety + ODD/CD, and 33% had ADHD alone.

Compared to children with ADHD alone, those with ADHD and anxiety did not show significant differences on any of the observed behaviors.  When ODD or CD was present in addition to ADHD, however, higher rates of interference, off-task behavior, verbal aggression to teacher, total aggression, and the ADHD composite were all significantly higher than for children with ADHD alone.
 

SUMMARY AND IMPLICATIONS

Results from this study reveal a wide range of differences in classroom behavior between children with and without ADHD.  Of particular interest were the clear differences observed between ADHD boys and girls.  ADHD girls had significantly lower scores than ADHD boys on the ADHD composite index.  They also showed less disruptive, rule-breaking behaviors than boys with ADHD, as exemplified by rates of interference and total aggression that were approximately half those of their male counterparts.

These findings have potentially important implications for identification and diagnosis.  Because girls with ADHD will generally be less overtly symptomatic, disruptive, and aggressive than boys with ADHD - even when they have been diagnosed with the combined subtype - they are likely to be experienced as less problematic by their teachers.  In fact, in addition to being less aggressive than boys with ADHD, girls with ADHD were no more aggressive or likely to be out of their seat than girls without ADHD. As a result, teachers may be less likely to pick up on these girls' difficulties.  In many instances, this could result in being identified much later than boys, or to never being identified at all.

The problem with this is that the longer ADHD goes untreated, the more likely it is to result in severely compromised academic performance as well as to the development of other difficulties.  This is likely to be compounded by the fact that the inattentive subtype of ADHD may be more common in girls than boys, and overt behavior problems are less pronounced in children with this subtype.  These observational findings highlight the need for teachers as well as clinicians to be aware of these gender-related differences and of their potential impact on the underidentification and misdiagnosis of girls with ADHD.

Results from the comorbidity analyses are also interesting.  What is clear from these results is that the presence of a disruptive behavior disorder is associated with even greater classroom behavioral impairment than is ADHD alone.  These findings highlight the importance of preventing the development of these other difficulties in children with ADHD.

In summary, results from this study provide the first observational evidence of importance behavioral differences between boys and girls with ADHD.  The nature of these differences is one that is likely to result in many cases of ADHD in girls being missed, or at least not diagnosed until substantially later than boys.  Unfortunately, this would increase the likelihood of significant academic problems and other emotional and/or behavioral difficulties developing as a girl's core ADHD symptoms go untreated.  Hopefully, information provided by studies such as this will reduced the frequency of such occurrences.

---

NEW INFORMATION ON GIRLS WITH ADHD

Despite a surge of interest in ADHD among girls, females are dramatically underrepresented in most studies of ADHD.  The extent to which prior research on ADHD has been conducted primarily with boys is difficult to overstate.  Several key studies that have provided a large amount of published data in the field did not include any girls, and in many studies where girls have been included, the number is typically too small to allow for the separate analysis and understanding of ADHD in girls.  The two largest studies of ADHD in girls are relatively recent and include the 116 females who participated in the MTA study and a second sample of 140 females with ADHD that has been conducted by a research group at Harvard (Biederman eta al., Clinical correlates of ADHD in females: Findings from a large group of girls ascertained from pediatric and psychiatric referral sources. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 526-533.)

Although these recent studies have provided valuable information on ADHD in girls, there are limitations to each.  In the MTA study, 75% of the girls were treated intensively with medication, behavioral treatments, or both for 14 months, and all displayed the combined subtype of ADHD.  Thus, results from this study are unlikely to be typical of how girls with ADHD appear in the general population.  In the study by the Harvard group, the sample was exclusively Caucasian and middle- to upper-middle class.  This restricted sample limits the extent to which results from this study can be generalized to the wider population of girls with ADHD.

The need for additional research on ADHD among girls is great, and a recently published study makes an important contribution to filling this prominent gap in the literature (Hinshaw, 2002. Preadolescent girls with AD/HD: I. Background characteristics, comorbidity, cognitive and social functioning, and parenting practices. Journal of Consulting and Clinical Psychology, 70, 1086-1097.)   Participants in this study were 245 6-12 year old girls with ADHD from the San Francisco Bay Area, and 88 comparison girls without ADHD recruited from the same communities.  Girls with ADHD were recruited from medical settings (e.g. pediatric practices, HMOs), mental health settings, school districts, ADHD parent groups, and newspaper advertisements.  Comparison girls were recruited from similar school districts, newspaper ads, and medical settings.  All girls in the ADHD group received a rigorous diagnostic evaluation - regardless of whether they had been previously been diagnosed - to insure that all met DSM-IV diagnostic criteria.  (For a review of current ADHD diagnostic criteria go to www.helpforadd.com/criteria-for-add/.  For information about evaluation guidelines, go to www.helpforadd.com/evaluation-guidelines/).

Among the participants, 53% were Caucasian, 27% were African American, 11% were Latina%, and 9% were Asian.  Girls from families across the entire socioeconomic spectrum - from families on public assistance to upper income families - were also represented.  Ninety-three of the girls with ADHD were diagnosed with the combined subtype (i.e. they showed both inattentive and hyperactive-impulsive symptoms), and 48 were diagnosed with the inattentive subtype (i.e. they showed predominantly inattentive symptoms and few if any hyperactive-impulsive symptoms).  Thus, this sample of girls with ADHD was more diverse than the samples that had previously been studied, and it is especially important that girls with the inattentive subtype of ADHD were included.  Relatively few girls meeting criteria for the hyperactive-impulsive subtype were found (this is the least frequently diagnosed ADHD subtype among school-age children), so these girls were not included in the analyses.

All girls - those with ADHD as well as the comparison girls - participated together in a 5-week summer enrichment day camp.  Daily activities included classroom, art, drama, and outdoor activities that allowed for ample social interaction and extensive observation of girls' behavior.  Classes of 25-26 girls (60% with ADHD and 40% comparison) participated together for each day's events.  Activities were supervised by a head teacher and 4-6 counselors who were unaware of which girls had been diagnosed with ADHD and which had not.  These staff provided daily ratings of the girls' behavior.  Parents of girls who were already taking medication were asked to have their daughters participated in the camp while unmedicated, and the majority complied with this request.

In order to take a comprehensive look at the characteristics of girls with ADHD, a wide range of information was collected including:

1) extensive demographic information on the girls and their families;

2) information on additional behavioral and emotional problems of the girls as assessed by a structured psychiatric interview and behavior rating scales;

3) daily behavior ratings conducted by counselors and other observers; As noted above, all observers were unaware of the girls' diagnostic status.

4) each girl's relationships with peers as measured through confidential peer interviews conducted at multiple times during the summer;

5) information on intellectual ability (IQ) and academic achievement measured by individually administered IQ and achievement tests;

6) parents' self-report of their parenting practices.  The different aspects of parenting assessed included: parental involvement, positive parenting, monitoring/supervision, discipline practices, and feelings of being overwhelmed in the parenting role.

These assessments make for a rich data set and provide the most comprehensive account obtained to date on the behavioral, emotional, social, academic, and family functioning among girls with ADHD.
 

RESULTS

The focus of data analysis was to examine differences between girls with ADHD, Combined Type (ADHD/C), girls with ADHD, Inattentive Type (ADHD/I), and comparison girls without ADHD (C) across multiple domains.  A summary of the results obtained is presented below. (Note: Because children with ADHD have increased rates of other psychiatric disorders, and these co-occurring problems could contribute to difficulties in multiple areas, the analyses conducted in this study specifically controlled for this possibility.  The results summarized below thus reflect differences between girls with ADHD and comparison girls after difficulties that could be explained by other disruptive behavior problems were taken into account).
 

Demographic factors

Factors examined here included age, annual family income, maternal education, ethnicity, and the percentage of girls in each group who came from two-parent households.  No significant differences between the 3 groups of girls were found.
 

Background and academic history

Characteristics considered here included the percentage of girls in each group who: 1) had been of low birth weight (< 2500 grams); 2) had a history of placement in special education services; 3) had repeated a grade; 4) had a history of speech/language problems; 4) had been adopted; and, 5) who had a documented history of abuse.  A number of group differences were found.

Relative to comparison girls, girls in both ADHD subgroups had significantly higher rates of special education placement (i.e. approximately 20% for both ADHD subtypes vs. 3.5% for comparison girls) and grade retention (ADHD/C - 14.1%; ADHD/I - 20.5%; C - 3.4%).  These findings highlight the substantially more difficult academic histories among girls with ADHD.

Girls with ADHD were also more likely to have experienced speech/language problems (ADHD/C - 25.6%; ADHD/I - 29.8%; C-7%), and to have been adopted (rates of 20.4%, 23.4%, and 4.5% respectively).  Girls with ADHD/C had also experienced higher rates of abuse than comparison girls (18.3% vs. 4.5%).  Rates of documented abuse in girls with ADHD/I were not elevated.
 

Comorbid psychiatric problems

The disorders examined were oppositional defiant disorder (ODD), conduct disorder (CD), anxiety disorders, depression, and reading disability.  (For a discussion of ODD and CD symptoms, go to www.helpforadd.com/co-occurring-disorders/).  Diagnoses were based on a structured psychiatric interview conducted with girls parents (i.e. the DISC -IV). Reading disability was defined as being in the bottom 15% on a standardized test of reading achievement.

As has been found in numerous studies of boys with ADHD, girls with ADHD were substantially more likely than comparison girls to have a host of additional problems.  The percentage of girls in each group diagnosed with each additional disorder is shown below.

ODD - ADHD/C - 71%; ADHD/I - 47%; C - 7%;

CD    - ADHD/C - 27%; ADHD/I - 11%; C - 0%;

Anxiety disorder - ADHD/C - 31%; ADHD/I - 19%; C - 3%;

Depression - ADHD/C - 10%; ADHD/I - 4%; C - 0%;

Reading Disability - ADHD/C - 11%; ADHD/I - 15%; C - 5%;
 

As can be seen, girls with both subtypes of ADHD were substantially more likely than comparison girls to be diagnosed with comorbid conditions.  In fact, the rates of co-occurring disruptive behavior disorders (i.e. ODD and CD) was equivalent to rates that have previously been reported in several samples of boys with ADHD.
 

Behavior Ratings

Behavior ratings on girls were obtained from parents, teachers (i.e. teachers from the summer camp program, not the girls' regular teachers), counselors, and other adults who observed the girls at camp.

As expected, girls in both ADHD groups received substantially higher ADHD symptom ratings from parents and teachers than comparison girls.  Rates of inattentive symptoms did not differ between girls in the two ADHD subgroups, suggesting equivalent levels of attention problems.

Parent and teacher ratings of oppositional behavior were highest for girls in the ADHD/C group.  Girls with inattentive ADHD, however, were also rated as displaying more oppositional behavior than comparison girls by parents and teachers.  Parents also rated both groups of girls with ADHD as displaying more symptoms of depression and emotional distress than comparison girls.

Differences in behavior ratings provided by observers were also evident.  Relative to comparison girls, girls in both ADHD groups were observed to engage in more frequent aggression towards peers (although the levels of overt aggression were low) and to be more non-compliant with teachers.
 

Self-reports of Emotional Distress

Girls also provide ratings of their own levels of depressive and anxiety symptoms.  Girls in both ADHD groups reported significantly more depressive symptoms than comparison girls, and did not differ from each other.  The average number of symptoms they reported, however, was still in a normal range.  No differences were found in the amount of anxiety symptoms that girls reported.
 

Peer relations

As noted above, all participants were interviewed several times during the summer about their relations with other girls at the camp.  In these interviews, girls were asked to identify the camp mates they liked most and the camp mates they liked least.  Girls with ADHD were significantly less likely than comparison girls to receive nominations for being liked most and significantly more likely to be nominated for "liked least".  Girls with the combined type of ADHD were especially likely to be "liked least" by their peers, while girls with ADHD/I were more likely to be seen as socially isolated.
 

Intellectual and Academic Functioning

IQ scores for girls with ADHD were significantly lower than scores obtained by comparison girls, although they fell exactly in the middle of the normal range.  Thus, although their IQ results were lower than scores obtained by girls without ADHD, they were not low in any absolute sense, and many girls in the ADHD groups obtained scores that were substantially above average.  No differences between girls in the different ADHD groups were found.

Achievement scores in reading and math were also lower for girls with ADHD than for comparison girls, but again fell right in the middle of the normal range.
 

Parenting practices

Self-reported parenting practices showed relatively few differences.  Thus, compared to parents of girls without ADHD, parents of girls with ADHD did not differ in their level of reported involvement, positive parenting, monitoring/supervision, or disciplinary practices.

Parents of girls with ADHD/C were more likely than parents of other girls to report endorsement of domineering parenting practices.  Not surprisingly, parents of girls with both types of ADHD were more likely to report feeling overwhelmed and less competent as a parent than parents of other girls.
 

SUMMARY AND CONCLUSIONS

Results from this study provide the most extensive information yet available on the functioning of girls with ADHD, and conclusively demonstrate that ADHD is far from a benign condition in girls.

In virtually every domain examined, girls with ADHD were experiencing significantly more difficulty than other girls.  Thus, girls with ADHD had substantially higher rates of other psychiatric problems, were more likely to have experienced academic difficulty (i.e. grade retention and special education placements), reported more distress, obtained lower scores on measures of cognitive functioning (although in the average range), and were less well liked by peers.  It should be emphasized that these differences were evident even after controlling for impairment that could be explained by higher levels of disruptive behavior problems among girls with ADHD, suggesting that ADHD is making a substantial contribution to these difficulties.

The fact that other important difficulties were often present in addition to ADHD highlights the need for comprehensive evaluation of girls' functionng in multiple domains, so that treatment plans which address the full scope of a child's difficulties can be developed.  Focusing strictly on whether or not a young girl has ADHD, and failing to consider the possibility of co-occurring problems that will also need to be addressed, is not sufficient.  The same holds true, of course, for evaluations conducted with boys suspected of having ADHD.

Relative to the consistent differences found between girls with both ADHD subtypes and comparison girls, differences between girls with ADHD/C and ADHD/I were not especially salient.  Exceptions to this were the higher rates of ODD and CD among girls with ADHD/C, their more frequent rejection by peers, and their more frequent history of documented abuse.  In contrast, girls with ADHD/I were more likely to be observed as showing social isolation.  Overall, however, girls with both types of ADHD showed fairly similar patterns and levels of impairment.

In conclusion, results from this important study highlight that girls with ADHD are likely to experience struggles in multiple domains on functioning.  In fact, although this study did not involve a direct comparison between boys and girls with ADHD, it appears that ADHD in girls is as impairing as it is among boys.  These results highlight the need for additional research on the processes and mechanisms by which ADHD leads to the development of multiple impairments in girls, as these developmental processes may differ from those that occur in boys.

In addition, it is imperative that parents, educators, and physicians become better informed about the manifestations of ADHD in girls, so that the greater tendency for girls to go undiagnosed and untreated is brought to an end.  The MTA study demonstrated that carefully conducted medication and behavioral treatments are effective for girls with ADHD as well as for boys, and making sure that a higher percentage of girls with ADHD receive such treatments is an important public health issue.