Health-Related Quality of Life in Children and Adolescents Who Have a

Diagnosis of Attention-Deficit/Hyperactivity Disorder

Anne F. Klassen, DPhil*; Anton Miller, MB, ChB, FRCPC*; and Stuart Fine, MB, FRCPC

ABSTRACT. Objective. The aim of treatment for at- cial domains include the following: role/social limita-
tention-deficit/hyperactivity disorder (ADHD) is to de- tions as a result of emotional-behavioral problems (REB),
crease symptoms, enhance functionality, and improve self-esteem (SE), mental health (MH), general behavior
well-being for the child and his or her close contacts. (BE), emotional impact on parent (PTE), and time impact
However, the measurement of treatment response is of- on parents (PTT). A separate domain measures limita-
ten limited to measuring symptoms using behavior rat- tions in family activities (FA). There is also a single-item
ing scales and checklists completed by teachers and par- measure of family cohesion (FC). Individual scale scores
ents. Because so much of the focus has been on symptom and summary scores for physical (PhS) and psychosocial
reduction, less is known about other possible health health (PsS) can be computed. Symptom severity data
problems, which can be measured easily using health- (parent and teacher) came from the Child/Adolescent
related quality-of-life (HRQL) questionnaires, which are Symptom Inventory 4. These checklists provide informa-
designed to gather information across a range of health tion on symptoms for the 3 ADHD subtypes (inattentive,
domains. The aim of our study was to measure HRQL in hyperactive, and combined). Each child underwent a
a clinic-based sample of children who had a diagnosis of comprehensive psychiatric assessment by 1 of 4 child
ADHD and consider the impact of 2 clinical factors, psychiatrists. Documentation included a full 5-axis Diag-
symptom severity and comorbidity, on HRQL. Our spe- nostic and Statistical Manual of Mental Disorders, Fourth
cific hypotheses were that parent-reported HRQL would Edition diagnosis on the basis of a comprehensive assess-
be poorer in children with ADHD than in normative US ment. Clinical information for each child was extracted
and Australian pediatric samples, in children with in- from hospital notes.
creasing severity of ADHD symptoms, and in children Results. Compared with both population samples,
who had diagnoses of comorbid psychiatric disorders. children with ADHD had comparable physical health
Methods. Cross-sectional survey was conducted in but clinically important deficits in HRQL in all psycho-
British Columbia, Canada. The sample included 165 re- social domains, FA, FC, and PsS, with effect sizes as
spondents of 259 eligible children (63.7% response rate) follows: FC 0.66, SE 0.90, MH 0.97, PTT
who were referred to the ADHD Clinic in British Colum- 1.07, REB 1.60, BE 1.73, PTE 1.87, FA
bia between November 2001 and October 2002. Children 1.95, and PsS 1.98. Poorer HRQL for all domains of
who are seen in this clinic come from all parts of the psychosocial health, FA, and PsS correlated significantly
province and are diverse in terms of socioeconomic status with more parent-reported inattentive, hyperactive, and
and case mix. ADHD was diagnosed in 131 children, combined symptoms of ADHD. Children with >2 comor-
68.7% of whom had a comorbid psychiatric disorder. bid disorders differed significantly from those with no
Some children had >1 comorbidity: 23 had 2, 5 had 3, and comorbidity in most areas, including RP, GH, REB, BE,
1 had 4. Fifty-one children had a comorbid learning dis- MH, SE, PTT, FA, and PsS, and from those with 1 comor-
order (LD), 45 had oppositional defiant disorder or con- bid disorder in 3 domains, including BE, MH, and FA
duct disorder (ODD/CD), and 27 had some other comor- and the PsS. The mean PsS score for children in the
bid diagnosis. The mean age of children was 10 years ODD/CD group (mean difference: 12.9; effect size
(standard deviation: 2.8). Boys composed 80.9% (N 106) 1.11) and children in the other comorbidity group (9.0;
of the sample. We used the 50-item parent version of the effect size .77) but not children in the LD group were
Child Health Questionnaire to measure physical and significantly lower than children with no comorbid dis-
psychosocial health. Physical domains include the fol- order. Predictors of physical health in a multiple regres-
lowing: physical functioning (PF), role/social limitations sion model included childs gender ( .177) and num-
as a result of physical health (RP), bodily pain/discom- ber of comorbid conditions ( .197). These 2
fort (BP), and general health perception (GH). Psychoso- variables explained very little variation in the PhS. Pre-
dictors of psychosocial health included the number of
comorbid conditions ( .374) and parent-rated com-
From the *Department of Pediatrics, University of British Columbia, and bined ADHD symptoms ( .362). These 2 variables
Center for Community Child Health Research, Childrens and Womens explained 31% of the variation in the PsS.
Health Center of British Columbia, Vancouver, British Columbia, Canada; Conclusions. Our study shows that ADHD has a sig-
and ADHD Clinic, Department of Psychiatry, Childrens and Womens nificant impact on multiple domains of HRQL in chil-
Health Center of British Columbia, and University of British Columbia, dren and adolescents. In support of our hypotheses, com-
Vancouver, British Columbia, Canada. pared with normative data, children with ADHD had
Accepted for publication Jun 30, 2004. more parent-reported problems in terms of emotional-
doi:10.1542/peds.2004-0844
behavioral role function, behavior, mental health, and
Reprint requests to (A.K.) Center for Community Child Health Research,
Childrens and Womens Health Center of British Columbia, 4480 Oak St,
self-esteem. In addition, the problems of children with
Vancouver, British Columbia, Canada V6H 3V4. E-mail: afk@ ADHD had a significant impact on the parents emo-
interchange.ubc.ca tional health and parents time to meet their own needs,
PEDIATRICS (ISSN 0031 4005). Copyright 2004 by the American Acad- and they interfered with family activities and family
emy of Pediatrics. cohesion. The differences that we found represent clini-

www.pediatrics.org/cgi/doi/10.1542/peds.2004-0844 PEDIATRICS Vol. 114 No. 5 November 2004 e541

Downloaded from by guest on November 28, 2016
cally important differences in HRQL. Our study adds In fact, current guidelines recommend the use of
new information about the HRQL of children with such scales to evaluate and monitor treatment ef-
ADHD in relation to symptom severity and comorbidity. fects.4 Because so much focus has been on measuring
Children with more symptoms of ADHD had worse psy- symptom reduction, little research has studied the
chosocial HRQL. Children with multiple comorbid dis- effects of ADHD on other aspects of a childs health.
orders had poorer psychosocial HRQL across a range of
However, the development of an increasing number
domains compared with children with none and 1 comor-
bid disorder. In addition, compared with children with
of multidimensional health-related quality-of-life
no comorbidity, psychosocial HRQL was significantly (HRQL) questionnaires now afford the opportunity
lower in children with ODD/CD and children in the to measure the health of children with ADHD more
other comorbidity group but not in children with an LD. comprehensively.58
The demonstration of a differential impact of ADHD on In the research literature, we are aware of 5 studies
health and well-being in relation to symptom severity that have measured HRQL in children with
and comorbidity has important implications for policies ADHD.913 Although these studies provide some in-
around eligibility for special educational and other sup- formation about the level of HRQL experienced by
portive services. Because the impact of ADHD is not children with ADHD, none have considered the re-
uniform, decisions about needed supports should incor- lationship between HRQL and key clinical factors,
porate a broader range of relevant indicators of outcome,
which we hypothesize will have an important effect
including HRQL. Although many studies focus on measur-
ing symptoms using rating scales and checklists, in our
on HRQL. The aim of our study was to measure
study, using a multidimensional questionnaire, we were HRQL in a clinic-based sample of children who have
able to show that many areas of health are affected in a diagnosis of ADHD and consider the impact of 2
children with ADHD. We therefore argue that research clinical factors on HRQL. Our specific hypotheses
studies of children with ADHD should include measure- were that parent-reported HRQL would be poorer in
ment of these broader domains of family impact and child children with ADHD than the general pediatric pop-
health. Pediatrics 2004;114:e541e547. URL: www.pediatrics. ulation, in children with increasing severity of
org/cgi/doi/10.1542/peds.2004-0844; attention-deficit/hyper- ADHD symptoms, and in children who have diag-
activity disorder, quality of life, comorbidity, children. noses of comorbid psychiatric disorders.

ABBREVIATIONS. ADHD, attention-deficit/hyperactivity disor- METHODS

der; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, Participants
Fourth Edition; ODD, oppositional defiant disorder; CD, conduct
disorder; LD, learning disability; HRQL, health-related quality of The sample was recruited from the ADHD clinic at British
life; CHQ, Child Health Questionnaire; CSI, Child/Adolescent Columbia Childrens Hospital, which is a specialist clinic located
Symptom Inventory 4; CHQ-PF50, 50-item parent-completed CHQ. at the only tertiary care pediatric facility in the province of British
Columbia (Canada). Children who are seen in this clinic come
from all parts of the province and are diverse in terms of socio-
economic status and case mix. Although many referred cases have

A
ttention-deficit/hyperactivity disorder (ADHD), complex diagnostic or management problems, approximately one
one of the most common childhood psychi- third of cases are newly identified by family doctors and are not
atric disorders, affects between 3% and 5% of necessarily complex.
children, according to the Diagnostic and Statistical A total of 335 children were referred to the ADHD clinic during
Manual of Mental Disorders, Fourth Edition (DSM-IV).1 our 12-month recruitment period (November 2001 to October
2002). Of these, 59 referrals were canceled, 8 parents could not
Boys are 3 times more likely to have ADHD than communicate in English, 5 children were redirected to a different
girls and 6 to 9 times more likely than girls to be seen clinic within the hospital, 3 children were involved in a different
with ADHD among clinic-referred children.2 ADHD research project, and 1 child died. Of the remaining 259 children,
is a condition characterized by different levels of 165 (63.7%) parents completed a study questionnaire. The final
study sample included 131 children who had a diagnosis of
inattention, hyperactivity, and impulsivity and gives ADHD (33 did not have ADHD, and for 1, the hospital notes were
rise to significant academic, social, and emotional missing). ADHD subtypes included the following: combined type
problems at home and at school.2 Academically, chil- (n 88); inattentive type (n 35); hyperactive (n 2); and not
dren with ADHD often underachieve or fail in otherwise specified (n 6). Family and child characteristics for the
sample appear in Table 1. Our study questionnaire was completed
school. Socially, they have poor relationships with by 116 (93.5%) biological parents, most commonly the childs
peers, teachers, and parents. Emotionally, they often mother (n 102; 82.3%). The mean age of children in the study
have poor self-esteem and are at considerably in- sample was 10 years (SD: 2.8). Boys composed 80.9% (n 106) of
creased risk for depression, anxiety, and/or delin- the sample.
quent behavior. The extent of comorbidity in chil-
dren with ADHD is high, as it commonly occurs in Materials
association with oppositional defiant disorder Before a childs appointment, parents are routinely sent a pack-
age of questionnaires to complete. For the purposes of our study,
(ODD), conduct disorder (CD), learning disabilities the Child Health Questionnaire (CHQ; described below) was in-
(LDs), and other psychiatric conditions such as anx- cluded in the package. Symptom severity data came from the
iety disorders and depression.2,3 Child/Adolescent Symptom Inventory 4 (CSI), one of the ques-
As with most mental health disorders in children, tionnaires sent routinely to parents (described below).
the aim of treatment for ADHD is to decrease symp-
toms, enhance functionality, and improve well-being CHQ
for the child and his or her close contacts. However, The CHQ is a multidimensional generic measure of HRQL that
can be used with children as young as 5 years.9 We used the
measurement of treatment response is often limited 50-item parent-completed CHQ (CHQ-PF50), which measures 11
to measuring a reduction in symptoms using behav- domains of health. Physical domains include the following: phys-
ior rating scales completed by teachers and parents. ical functioning, role/social limitations as a result of physical

e542 HEALTH-RELATED QUALITY OF LIFE IN CHILDREN WITH ADHD

Downloaded from by guest on November 28, 2016
TABLE 1. Demographic, Socioeconomic, and Clinical Charac- the head of the ADHD clinic describing the study, a consent form,
teristics of the Family and the Child and the study questionnaire were sent to all new referrals over a
12-month period. A reminder letter and up to 2 additional copies
Characteristic n % of the questionnaires were sent to nonrespondents. Remaining
Relationship to child nonrespondents were contacted by telephone to confirm response
Biological parent 116 93.5 status. Each child underwent a comprehensive psychiatric assess-
Foster parent 4 3.2 ment by 1 of 4 child psychiatrists, all of whom have a minimum of
Adoptive parent 3 2.4 10 years of clinical and research experience with ADHD. Diagno-
Guardians 1 0.8 sis was based on LEAD (longitudinal, expert, all data) criteria.21
Marital status Psychoeducational and occupational therapy testing was done as
Married 80 61.5 needed. Documentation included a full 5-axis DSM-IV diagnosis
Common law 9 6.9 based on a comprehensive assessment. Clinical information for
Single 11 8.5 each child was extracted from hospital notes.
Separated 13 10.0
Divorced 17 13.1 Data Analysis
Education To explore relationships between symptom severity scores and
Some high school 15 11.5 comorbidity by gender and age, we used 2 tests, t tests, and
High school diploma 22 16.9 Mann-Whitney U tests. Interclass correlation coefficients and
Some postsecondary 21 16.2 paired t tests were computed to examine agreement between
Diploma from trade school or college 48 36.9 parent and teacher CSI data. Mean scores for the CHQ-PF50 were
or some university education computed and compared with published population norms. Be-
University degree 24 18.5 cause no population norms are available for Canadian children,
Household income, n (%) comparisons were made with US9 and Australian14 norms. Z tests
$20 000 23 18.7 were used to estimate statistical significance, and effect sizes were
$20 000$40 000 32 26.0 computed to explore the clinical importance of differences be-
$40 000$60 000 27 21.9 tween the ADHD sample and the US norms (Australian SDs were
$60 000$80 000 12 9.8 not available). Effect sizes were computed by dividing the mean
$80 000 29 23.6 difference in scores by the population SD. Cohens guidelines for
Gender interpretation (0.2 is small, 0.5 is medium, 0.8 is large) were
Male 106 80.9 used.22 Spearman correlation coefficients were used to examine
Female 25 19.1 the relationship between HRQL and ADHD symptom severity. A
Child age t test, Mann Whitney U test, or analysis of variance was used to
67 31 23.7 compare mean CHQ-PF50 scores by gender and comorbidity.
89 36 27.5 Effect sizes for these were computed, using the pooled SD. Pear-
1011 24 18.3 son correlation coefficients identified significant relationships be-
1213 21 16.0 tween CHQ-PF50 summary scores and the following variables:
1415 14 10.7 annual household income, caregiver education, marital status,
1617 5 3.8 child age, child gender, type of ADHD, the presence of comorbid-
Type of ADHD ity, type of comorbidity, number of comorbid diagnoses, and
Combined 88 67.2 parent- and teacher-reported ADHD symptoms. Variables that
Inattentive 35 26.7 were significantly related to the summary scores were entered into
Not otherwise specified 8 6.1 multiple regression models to examine the proportion of variation
Type of comorbid psychiatric diagnosis explained in HRQL summary scores.
LD 51 38.9
ODD/CD 45 34.4 RESULTS
Other 27 20.6 Treatment History
Psychosocial stressors 48 36.6
A history of stimulant medication was noted for 59
(45%) children. Ritalin had been prescribed to 54 and
Dexedrine to 20 (16 of whom had also had Ritalin).
health, bodily pain/discomfort, and general health perception.
Psychosocial domains include the following: role/social limita-
At the time of their appointment, 92 (70.2%) children
tions as a result of emotional-behavioral problems, self-esteem, were not on medication, 36 (27.5%) were receiving
mental health, general behavior, emotional impact on parent, and medication, and for 3 (2.3%) children, the medication
time impact on parents. A separate domain measures limitations status was unknown. Just over half of the children on
in family activities. There is also a single-item measure of family medication for ADHD had their medication changed
cohesion. Scores for the domains and single item range from 0 to
100, with higher scores indicating better HRQL. In addition, sum- or adjusted by the psychiatrist.
mary scores can be derived to measure physical and psychosocial
health. The summary scores have a mean of 50 and an SD of 10. Comorbid Diagnoses
The CHQ-PF50 has undergone extensive validation, and norma- Ninety (68.7%) children had at least 1 comorbid
tive data are available.9,14 psychiatric diagnosis. Some children (n 29; 22.2%)
had 1 comorbidity: 23 (17.6%) had 2, 5 (3.8%) had 3,
CSI
and 1 (0.8%) had 4. Fifty-one (38.9%) children had a
These checklists provide information on symptoms for the 3
ADHD subtypes: 9 inattentive and 9 hyperactivity/impulsivity
comorbid LD, 45 (34.4%) had ODD/CD, and 27
symptoms are listed in 2 separate scales, and both scales together (20.6%) had some other comorbid diagnosis (eg,
form the combined subtype.1518 The symptom severity approach Tourettes syndrome, sleep disorder, depression,
sums the values of the items (never 0, sometimes 1, anxiety disorder, communication disorder). The pro-
often 2, very often 3). The instrument has moderate to portion of children with versus without a comorbid
high sensitivity for DSM-IV disorders and correlates well with
dimensional rating scales such as the Achenbach19 and semistruc- diagnosis did not vary by gender and was not related
tured interviews.20 to age.

Procedure ADHD Symptom Severity

Ethical approval was obtained from the University of British Parent-reported CSI data were available for 122
Columbia and the participating childrens hospital. A letter from (93.1%) children, and teacher-reported CSI data were

www.pediatrics.org/cgi/doi/10.1542/peds.2004-0844 e543
Downloaded from by guest on November 28, 2016
available for 119 (90.8%) children. Interclass correla- was associated with poorer HRQL for all domains of
tion coefficients computed for parent- and teacher- psychosocial health, family activities, and the psy-
reported CSI scores were moderate to low in magni- chosocial summary score (see Table 2). Similar cor-
tude for ADHD inattentive subtype (0.34; P .027; relations between teacher-reported scores (CSI) and
n 92), ADHD hyperactive subtype (0.39; P .011; the psychosocial HRQL scores were not statistically
n 87), and ADHD combined subtype (0.30; P significant (all correlations below r 0.14).
.069; n 72). Parents rated ADHD symptoms signif-
icantly higher than teachers for ADHD inattentive HRQL by Number of Comorbid Psychiatric Diagnoses
subtype (6.63 vs 5.89; t 2.034, df 91, P .045) and
Table 3 shows the mean CHQ-PF50 item, domain
ADHD combined subtype (11.60 vs 9.86; t 2.433,
and summary scores, and 95% confidence intervals
df 71, P .018).
by number of comorbid disorders (0, 1, 2). Poorer
HRQL of Children With ADHD Compared With health for those with more comorbidity was appar-
Population Norms ent for 5 domains of psychosocial health, 2 domains
of physical health, family activities, and the psycho-
Mean CHQ-PF50 scores for the ADHD sample and
social summary score. The differences in mean scores
Australian and US population norms appear in Fig 1.
comparing those with 2 comorbidities and those
The physical health domains for children with
with no comorbidity resulted in large effect sizes (see
ADHD did not vary from either normative sample.
Table 3). No differences were detected when the
All psychosocial health domains, family activities,
group with no comorbidity was compared with the
family cohesion, and the psychosocial summary
group with 1 comorbid disorder. Those with 2
score were substantially lower in children with
comorbid disorders differed significantly from those
ADHD compared with the 2 population samples (all
with no comorbidity in most areas, including role
ADHD vs US; P .0001).
physical, general health, role emotion/behavioral,
To estimate the clinical importance of these differ-
behavior, mental health, self-esteem, parental im-
ences, effect sizes were computed between the
pact-time, family activities, and the psychosocial
ADHD and US samples. The results indicate clini-
summary score, and from those with 1 comorbid
cally important deficits in HRQL in all psychosocial
disorder in 3 domains, including behavior, mental
domains, family activities, family cohesion, and the
health, family activities, and the psychosocial sum-
psychosocial summary score and in order of magni-
mary score.
tude were as follows: family-cohesion 0.66, self-
esteem 0.90, mental health 0.97, parental
impact-time 1.07, role/social limitations emo- HRQL by Type of Comorbid Psychiatric Diagnoses
tional-behavioral 1.60, general behavior The physical and psychosocial CHQ-PF50 sum-
1.73, parental impact-emotional 1.87, family mary scores for children with no comorbid diagnoses
activities 1.95, and psychosocial summary score were compared with children with an LD, ODD/CD,
1.98. or other comorbid disorder. Although no differences
were found for physical health, the mean score for
Correlations Between HRQL and Symptoms of ADHD psychosocial health for children in the ODD/CD
Spearman correlations between parent-reported group (mean difference: 12.9; P .001; effect size
CSI scores and HRQL physical health domains were 1.11) and children in the other comorbidity group
not significant. Higher ADHD symptom severity (9.0; P .005; effect size .77) but not children in

Fig 1. Mean CHQ-PF50 item and domain scores

comparing Australian and US population norms
with ADHD children. PF indicates physical func-
tion; RP, role physical; BP, bodily pain; GH, general
health; REB, role emotional-behavioral; BE, behav-
ior; MH, mental health; SE, self-esteem; PE, parental
impact-emotional; PT, parental impact-time; FA,
family activities; FC, family cohesion.

e544 HEALTH-RELATED QUALITY OF LIFE IN CHILDREN WITH ADHD

Downloaded from by guest on November 28, 2016
 TABLE 2. Spearman Correlations Between Parent-Reported CSI Symptom Severity Score Scales
and CHQ-PF50 Item, Domain, and Summary Scores
ADHD Inattentive ADHD Hyperactive ADHD Combined
Subtype Subtype Subtype
Physical function .09 .16 .13
Role physical .01 .17 .12
Bodily pain .14 .06 .13
General health .10 .04 .11
Role emotional-behavioral .28* .31* .35*
Behavior .34* .39* .48*
Mental health .26* .23 .31*
Self-esteem .34* .06 .21
Parental impact-emotional .39* .28* .42*
Parental impact-time .32* .33* .37*
Family activities .35* .32* .38*
Family cohesion .17 .02 .09
Physical summary score .05 .02 .00
Psychosocial summary score .47* .35* .48*
* Correlation is significant at the 0.01 level (2-tailed).
Correlation is significant at the 0.05 level (2-tailed).

TABLE 3. Mean CHQ-PF50 Item, Domain and Summary Scores, 95% Confidence Intervals, P Values, and Effect Sizes for Children by
Number of Comorbid Psychiatric Diagnoses
0, Mean 1, Mean 2, Mean P Value Effect Size
(SD; n 41) (SD; n 61) (SD; n 29)
Physical function 96.8 (11.0) 98.1 (5.6) 94.0 (17.9) .313 .25
Role physical 97.1 (9.2) 98.9 (4.2) 88.5 (24.4) .030 .64
Bodily pain 84.8 (22.2) 83.8 (17.0) 75.9 (26.9) .299 .42
General health 78.3 (15.1) 76.8 (15.6) 65.9 (14.4) .002 .79
Role emotional-behavioral 71.9 (30.1) 63.2 (34.4) 49.0 (35.9) .022 .67
Behavior 52.4 (17.8) 48.6 (20.8) 34.5 (16.6) .001 .90
Mental health 71.2 (14.5) 66.3 (17.2) 56.7 (19.9) .003 .82
Self-esteem 71.6 (16.3) 63.7 (23.9) 54.4 (16.5) .003 .82
Parental impact-emotional 48.6 (20.7) 44.9 (20.0) 37.9 (20.6) .101 .52
Parental impact-time 73.1 (22.8) 66.7 (22.3) 57.1 (27.5) .029 .66
Family activities 63.2 (26.4) 55.4 (28.1) 35.6 (26.9) .001 .95
Family cohesion 62.9 (26.6) 58.3 (28.9) 50.5 (26.0) .182 .45
Physical summary score 56.6 (5.8) 57.1 (5.1) 52.8 (11.6) .262 .51
Psychosocial summary score 38.2 (9.1) 33.4 (12.9) 26.4 (10.4) .001 .98

the LD group were significantly lower than children conditions ( .197; P .028). These 2 variables
with no comorbid disorder. explained very little variation in physical health.
Some children in each of the above groups had 1 Predictors that were significant in the final model
type of comorbid disorder. We therefore compared (n 93) for psychosocial health included the number
the psychosocial summary score for smaller groups of comorbid conditions ( .374; P .001) and
of children with specific patterns of comorbid disor- parent-rated combined ADHD symptoms (
ders to examine the relationship with HRQL. The .362; P .001). In the final model, the adjusted R2
subgroups included children with only an ODD/CD was .31 (F 32.051, df 2,90, P .0001).
(n 21); only an LD (n 28); only 1 of the other
comorbidities (n 12); both an ODD/CD and an LD DISCUSSION
(n 14); and the remaining children (n 15), who In agreement with previous research,912 our study
had a mix of 2 or more psychiatric disorders. Chil- shows that ADHD has a significant impact on mul-
dren with an ODD/CD scored lower than children tiple domains of HRQL in children and adolescents.
with no comorbidity (mean difference: 13.0; P Specifically, in support of our first hypothesis, com-
.001) and those with an LD (13.4; P .001). Chil- pared with normative data, children with ADHD
dren with ODD/CD and an LD had lower mean had more parent-reported problems in terms of emo-
scores than those with no comorbidity (10.2; P tional-behavioral role function, behavior, mental
.024) and those with an LD (10.6; P .025). Finally, health, and self-esteem. In addition, the problems of
children in the mixed group had lower mean scores children with ADHD had a significant impact on the
than those with no comorbidity (13.4; P .001) and parents emotional health and parents time to meet
those with an LD (13.9; P .002). their own needs, and they interfered with family
activities and family cohesion. No differences were
Predictors of HRQL found for aspects of physical health. Researchers
Predictors that were significant in the final model have reached a consensus that a minimally impor-
(n 121) for physical health containing only the tant difference in HRQL is close to one half of an
significant bivariate predictors included childs gen- SD.23 The differences that we found (eg, effect size of
der ( .177; P .049) and number of comorbid 1.98 for the psychosocial summary score) were sub-

www.pediatrics.org/cgi/doi/10.1542/peds.2004-0844 e545
Downloaded from by guest on November 28, 2016
stantially larger and therefore represent clinically im- outcome measures that are sensitive to the impact
portant differences in HRQL. that symptom severity and comorbid disorders have
Our study adds new information about the HRQL on the overall health and well-being of children. The
of children with ADHD for 2 clinical characteristics. collection of such information could provide useful
First, as hypothesized, HRQL was correlated with information to parents and children who may want
parent-reported ADHD symptom severity: children to know the likely impact that their childs psychiat-
with more ADHD symptoms had poorer HRQL. Par- ric diagnoses will have on his or her health, function,
ent-reported ADHD symptoms were an important and family life. Our results suggest that whereas the
predictor of psychosocial health in the regression addition of an LD will not make much difference, the
model. These findings are a graphic illustration of addition of an ODD/CD will have clinically impor-
how ADHD affects health for the child and the fam- tant implications for the child and his or her family.
ily in ways that are broader than is apparent from The demonstration in our findings of a differential
symptoms alone and that are measurable with ap- impact of ADHD on health and well-being in relation
propriate instruments. Although many treatment to symptom severity and comorbidity has important
studies focus simply on measuring a reduction in implications for policies around eligibility for special
symptoms using rating scales and checklists, we ar- educational and other supportive services to children
gue that measurement of these broader domains of and youths with developmental and behavioral dis-
family impact and child health be included in clinical orders. Our findings illustrate a serious flaw in tying
and research studies of children with ADHD. In our service provision to specific medical diagnosis, as
study, using a multidimensional questionnaire, we happens in many parts of Canada and the United
were able to show that ADHD symptoms are related States. It is clear that the impact of ADHD is not
to many areas of health of children and their family. uniform, and, hence, decisions about needed sup-
The finding that parent and teacher ratings of ports should incorporate a broader range of function-
ADHD symptoms were poorly correlated is in ally relevant indicators.
agreement with other research on this topic. Many There are certain limitations to this study. We were
researchers have noted that parents and teachers not able to compare participants and nonparticipants
frequently disagree on their assessment of behavior- in terms of clinical variables and therefore do not
al/emotional problems in children.19,2426 Such dif- know the possible direction of bias, if any. We cannot
ferences do not mean that either reporter is inaccu- discount the possibility that the low level of HRQL
rate, because parents and teachers see the child in that we found could be attributable to a poorer level
different situations and ratings may be affected by of HRQL in our clinic-based sample of children com-
many different factors. Often teachers see the chil- pared with children with ADHD in the general com-
dren only when they are medicated (in Canada, long- munity. Our comparisons with population norms for
acting stimulants have been available only recently), the CHQ were limited by not having Canadian
which could help to account for the finding that norms, and we recognize that cross-country compar-
HRQL was related to parent-reported symptoms but isons are limited by differences between samples in
not teacher-reported symptoms. It is also possible factors such as socioeconomic status, age, gender,
that parents may have exaggerated both their childs and health care system. NonEnglish-speaking fam-
ADHD symptoms and HRQL in an attempt to secure ilies were excluded from the study, although our
an appointment. numbers of these were small. HRQL and ADHD
Second, in support of our hypothesis, an important symptom data both were provided by parents and
predictor of psychosocial HRQL was the presence of therefore represent only 1 perspective. This may ac-
comorbid psychiatric disorders. We found that chil- count for the strong correlation between HRQL and
dren with 2 comorbid disorders had poorer psy- parent- but not teacher-rated symptoms. Parents
chosocial HRQL in a range of domains compared were used as proxies, although research shows that
with children with none and 1 comorbid disorder. parent and child often differ in their ratings of child
Furthermore, at the bivariate level, HRQL was re- HRQL.28,29 Given the nature of ADHD (ie, children
lated to the particular type of comorbidity. Com- are inattentive, hyperactive, and impulsive, and their
pared with children with no comorbidity, psychoso- parents have a higher prevalence of psychopatholo-
cial HRQL was significantly lower in children with gy30,31), research into the validity of self- and proxy
an ODD/CD and children in the other comorbidity report is called for. Finally, because so little is known
group but not children with an LD. The effect sizes about the HRQL of children with ADHD, we have
for these differences were large, indicating clinically provided a comprehensive description of the study
important deficits in health. When we separated chil- sample. Because this necessitated multiple statistical
dren into 6 subgroups by comorbidity, we found that comparisons, the possibility of an error cannot be
compared with children with no comorbidity and discounted.
children with an LD, psychosocial HRQL was lower
in children with only ODD/CD, ODD/CD and LD, CONCLUSION
and a mix of comorbidity conditions. Children with ADHD were found to have substan-
Given that comorbidity is common with tially lower HRQL compared with normative data.
ADHD2,3,27 and comorbidity is an important predic- Psychosocial health was related to ADHD symptoms
tor of HRQL in ADHD, the effectiveness of interven- and number of comorbid disorders. HRQL is an
tions in ADHD, if they are aimed at improving over- important outcome that has received little attention
all health and functioning, should be evaluated using in children with ADHD. The use of tools such as the

e546 HEALTH-RELATED QUALITY OF LIFE IN CHILDREN WITH ADHD

Downloaded from by guest on November 28, 2016
CHQ shows how additional and useful information 14. Waters E, Salmon L, Wake M. The parent-form Child Health Question-
naire in Australia: comparison of reliability, validity, structure, and
that is relevant to the life of the patient and his or her norms. J Pediatr Psychol. 2000;25:381391
family can be obtained. 15. Gadow KD, Sprafkin J. Child Symptom Inventory 4 Norms Manual. Stony
Brook, NY: Checkmate Plus LTD; 1997
16. Gadow KD, Sprafkin J. Child Symptom Inventory 4 Screening Manual.
ACKNOWLEDGMENTS Stony Brook, NY: Checkmate Plus LTD; 1998
This study was funded by a grant from the British Columbia 17. Gadow KD, Sprafkin J. Adolescent Symptom Inventory 4 Norms Manual.
Medical Services Foundation. Dr Klassen holds a Senior Research Stony Brook, NY: Checkmate Plus LTD; 1998
Fellowship from the Canadian Institutes of Health Research. 18. Gadow KD, Sprafkin J. Adolescent Symptom Inventory 4 Screening Manual.
Stony Brook, NY: Checkmate Plus LTD; 1997
19. Achenbach TM, McConaughy SH, Howell CT. Child/adolescent behav-
REFERENCES ioral and emotional problems: implications of cross-informant correla-
tions for situational specificity. Psychol Bull. 1987;101:213232
1. American Psychiatric Association. Diagnostic and Statistical Manual of 20. Grayson P, Carlson GA. The utility of a DSMIIIR based checklist in
Mental Disorders. 4th ed. Washington, DC: American Psychiatric screening child psychiatric patients. J Am Acad Child Adolesc Psychiatry.
Association; 1994 1991;30:669 673
2. Barkley RA. Attention-Deficit Hyperactivity Disorder. A Handbook for Di- 21. Spitzer RL. Psychiatric diagnosis: are clinicians still necessary? Compr
agnosis and Treatment. 2nd ed. New York, NY: Guilford Press; 1998 Psychiatry. 1983;24:399 411
3. Biederman J, Newcorn J, Sprich S. Comorbidity of attention deficit 22. Cohen J. Statistical Power for the Behavioural Sciences. New York, NY:
hyperactivity disorder with conduct, depressive, anxiety, and other Academic Press; 1977
disorders. Am J Psychiatry. 1991;148:564 577 23. Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in
4. Committee on Quality Improvement, Subcommittee on Attention- health-related quality of life: the remarkable universality of half a
Deficit/Hyperactivity Disorder. Clinical practice guideline: diagnosis standard deviation. Med Care. 2003;41:582592
and evaluation of the child with attention-deficit/hyperactivity disor- 24. Kolko DJ, Kazdin AE. Emotional/behavioral problems in clinic and
der. Pediatrics. 2000;105:1158 1170 nonclinic children: correspondence among child, parent and teacher
5. Eiser C, Morse R. A review of measures of quality of life for children report. J Child Psychol Psychiatry. 1993;34:9911006
with chronic illness. Arch Dis Child. 2001;84:205211 25. Kumpulainen K, Rasanen E, Henttonen I, et al. Childrens behavioral/
6. Schmidt LJ, Garrat AM, Fitzpatrick R. Child/parent-assessed popula- emotional problems: a comparison of parents and teachers reports for
tion health outcome measures: a structured review. Child Care Health elementary school-aged children. Eur Child Adolesc Psychiatry. 1999;
Dev. 2002;28:227237 8(suppl 4):41 47
7. Spieth LE, Harris CV. Assessment of health related quality of life in 26. Mitsis EM, McKay KE, Schulz KP, Newcorn JH, Halperin JM. Parent-
children and adolescents: an integrative review. J Pediatr Psychol. 1996; teacher concordance for DSM-IV attention-deficit/hyperactivity disor-
21:175193 der in a clinic-referred sample. J Am Acad Child Adolesc Psychiatry.
8. Connolly MA, Johnson JA. Measuring quality of life in paediatric pa- 2000;39:308 313
tients. Pharmacoeconomics. 1999;16:605 625 27. Kutcher S, Aman M, Brooks SJ, et al. International consensus statement
9. Landgraf JM, Abetz L, Ware JE. The CHQ Users Manual. Boston, MA: on attention-deficit/hyperactivity disorder (ADHD) and ODD/CDs
Health Act; 1999 (DBDs): clinical implications and treatment practice suggestions. Eur
10. Sawyer MG, Whaites L, Rey JM, Hazell PL, Graetz BW, Baghurst P. Neuropsychopharmacol. 2004;14:1128
Health-related quality of life of children and adolescents with mental 28. Eiser C, Morse R. Quality-of-life measures in chronic diseases of child-
disorders. J Am Acad Child Adolesc Psychiatry. 2002;41:530 537 hood. Health Technol Assess. 2001;5:55 62
11. DeVeaugh-Geiss J, Conners CK, Sarkis EH, et al. GW320659 for the 29. Waters E, Stewart-Brown S, Fitzpatrick R. Agreement between adoles-
treatment of attention-deficit/hyperactivity disorder in children. J Am cent self-report and parent reports of health and well-being: results of
Acad Child Adolesc Psychiatry. 2002;41:914 920 an epidemiological study. Child Care Health Dev. 2003;29:501509
12. Bastiaansen D, Koot HM, Ferdinand RF, Verhulst FC. Quality of life in 30. Barkley RA, DuPaul GJ, McMurray MB. Comprehensive evaluation of
children with psychiatric disorders; self-, parent, and clinician report. attention deficit disorder with and without hyperactivity as defined by
J Am Acad Child Adolesc Psychiatry. 2004;43:221230 research criteria. J Consult Clin Psychol. 1990;58:775789
13. Landgraf JM, Rich M, Rappaport L. Measuring quality of life in children 31. Biederman J, Faraone SV, Keenan K, et al. Further evidence for family-
with attention-deficit/hyperactivity disorder and their families: devel- genetic risk factors in attention deficit hyperactivity disorder. Patterns
opment and evaluation of a new tool. Arch Pediatr Adolesc Med. 2002; of comorbidity in probands and relatives psychiatrically and pediatri-
156:384 391 cally referred samples. Arch Gen Psychiatry. 1992;49:728 738

www.pediatrics.org/cgi/doi/10.1542/peds.2004-0844 e547
Downloaded from by guest on November 28, 2016
 Health-Related Quality of Life in Children and Adolescents Who Have a
Diagnosis of Attention-Deficit/Hyperactivity Disorder
Anne F. Klassen, Anton Miller and Stuart Fine
Pediatrics 2004;114;e541
DOI: 10.1542/peds.2004-0844
Updated Information & including high resolution figures, can be found at:
Services /content/114/5/e541.full.html
References This article cites 22 articles, 4 of which can be accessed free
at:
/content/114/5/e541.full.html#ref-list-1
Citations This article has been cited by 24 HighWire-hosted articles:
/content/114/5/e541.full.html#related-urls
Subspecialty Collections This article, along with others on similar topics, appears in
the following collection(s):
Developmental/Behavioral Pediatrics
/cgi/collection/development:behavioral_issues_sub
Attention-Deficit/Hyperactivity Disorder (ADHD)
/cgi/collection/attention-deficit:hyperactivity_disorder_adhd_
sub
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
/site/misc/reprints.xhtml

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright 2004 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from by guest on November 28, 2016

 Health-Related Quality of Life in Children and Adolescents Who Have a
Diagnosis of Attention-Deficit/Hyperactivity Disorder
Anne F. Klassen, Anton Miller and Stuart Fine
Pediatrics 2004;114;e541
DOI: 10.1542/peds.2004-0844

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
/content/114/5/e541.full.html

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2004 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from by guest on November 28, 2016