ORI GIN AL PA PER

Improvement in Global Psychopathology IncreasesQuality of Life During Treatment of ADHDwith Atomoxetine or Stimulants

Leo Bastiaens

Published online: 25 March 2011� Springer Science+Business Media, LLC 2011

Abstract To evaluate what determines the increase in quality of life during treatment for

ADHD: improvement in core ADHD symptoms or improvement in global psychopathol-

ogy ratings. A prospective follow-up of ADHD patients in one community clinic. Stan-

dardized evaluation and outcome measures were used, including the Mini International

Neuropsychiatric Interview, Child Symptom Inventory, 18 item ADHD rating scale, and

the Health and Life Functioning Scale. 75 patients between the ages of 6 and 12 were

treated with atomoxetine or stimulants with a stable dose for 10 months. At end point,

there were modest improvements in ADHD symptoms, global psychopathology, level of

functioning and quality of life. The improvement in quality of life was driven by a decrease

in global psychopathology, not by a decrease in ADHD symptoms. The treatment for

ADHD may need to be broadened beyond the core symptoms. A chronic disease man-

agement model may well be applicable.

Keywords ADHD � Quality of life � Psychopathology

Introduction

Attention deficit hyperactivity disorder (ADHD) is a chronic and impairing condition,

significantly impacting on quality of life and creating a substantial economic burden [1, 2].

Co-morbidity in ADHD appears to be the norm, rather than the exception [3]. How this

co-occurrence with other symptoms and syndromes impact on ADHD outcomes, such as

quality of life, is not very clear yet [4, 5]. In general, the presence of co-morbidity may lead

to more psychosocial and emotional difficulties in the future [6]. Treatment outcomes may

differ between groups of ADHD patients as well. For example, differential response to

ADHD treatment, based on the presence of internalizing and/or externalizing co-morbid

problems, was apparent in the MTA study [7]. Another finding of the MTA study was the

inferior outcome of routine community care, compared to the research protocols [8].

L. Bastiaens (&)University of Pittsburgh, 33 Sunnyhill Drive, PA, Pittsburgh 15228, USAe-mail: bastiaensl@aol.com

123

Psychiatr Q (2011) 82:303–308DOI 10.1007/s11126-011-9172-4

In general, outcomes in community care are modest at best [9]. How co-morbidity

impacts on this modest response to community care for ADHD is not well known. A study

in 2008 found equal improvements in quality of life in community clinic patients treated

with atomoxetine or stimulants over a period of 8 months [10]. Age, participation in

psychotherapy, and parental disability were not correlated with outcome, but the impact of

co-morbidity was not examined. In that study, the increase in quality of life was only 16

percent. The possibility exists that co-morbidity is higher in community samples than in

research subjects, thereby reducing the overall response. If this is the case, then additional

treatments, targeting co-morbid problems, may be necessary to improve results in com-

munity care.

As a first step to investigate the role of co-morbid issues, the current study evaluated

what determines the increase in quality of life during community treatment of ADHD with

atomoxetine or stimulants: the improvement in core ADHD symptoms or the improvement

in a global measure of psychopathology, a measure indicative of internalizing and exter-

nalizing co-existing symptoms.

Methods

Patients, between 6 and 12 years of age, whose primary diagnosis was ADHD, with or

without other conditions, were included in this study. Patients with a primary diagnosis

requiring other pharmacotherapies, such as depression, anxiety, bipolar disorder, or per-

vasive developmental disorder, albeit co-morbid with ADHD, were not included. A psy-

chiatric evaluation by a board certified child and adolescent psychiatrist in this clinic

includes the administration of the Mini International Neuropsychiatric Interview for

Children and Adolescents [11] to the patient, and the completion of the Child Symptom

Inventory-IV (CSI; [12]) by a parent.

The CSI measures 92 internalizing and externalizing symptoms on a scale from 0

(never) to 3 (very often), although some symptoms are rated in a yes/no fashion. It includes

the 18 item ADHD rating scale, ranging from 0 to 54. The most extensive scales, within the

CSI, measure symptoms related to oppositional defiant disorder, conduct disorder, gen-

eralized, social, and separation anxiety disorder, depressive disorders, and pervasive

developmental disorders, while other screening questions address other anxiety disorders,

tics, and elimination problems. The total score ranges from 0 to 281. In this study, the CSI

total score was considered an index of global psychopathology.

During the initial assessment, the parent filled out the Health and Life Functioning Scale

(HALFS). The HALFS is a 10 item, parent-rated, quality-of-life scale, indicating func-

tioning in general health, academics, leisure activities, family relations, and social arena.

Its score ranges from 0 to 20, with 20 indicating excellent quality of life. The HALFS was

found to be a reliable, valid, and responsive quality-of-life scale for use in pediatric

psychiatric patients [13].

In general, patients were started on pharmacotherapy, if indicated, immediately fol-

lowing the assessment, were scheduled for a return visit 1 month later, and were subse-

quently seen for follow-up every 2–3 months. Patients were treated with atomoxetine or

stimulants with dose titrations on clinical grounds. The decision to treat with atomoxetine

or stimulants was made with informed consent/assent of the parent and the patient, and

included issues such as individual preference, formulary status, and prior exposure to

treatment. Many patients were also encouraged to start psychotherapy, but only 45%

(34/75) received 8 or more sessions at the time of the endpoint evaluation.

304 Psychiatr Q (2011) 82:303–308

123

Between September 2009 and September 2010, the CSI and the HALFS were repeated

by the parent, and the Global Assessment of Functioning (GAF) was determined by the

treating psychiatrist, as long as the patient had been taken the same medication and dose

for at least three consecutive months. This time frame was chosen because a previous study

in the same clinic showed that quality-of-life improvement levels off after 3 months,

indicating stability in improvement between 3 and 12 months of treatment [9]. The GAF

was rated without knowledge of the Child Symptom Inventory or HALFS.

Statistics

Statistical comparisons were done with Student’s t-test for continuous variables. Spearman

rank was used to test for correlations. Tests were two-tailed and significance was put at the

0.05 level. Two multiple regression analyses were performed with the HALFS and GAF as

dependent variables, while both the 18 item ADHD rating scale score and the total CSI

score were entered as independent variables.

Results

Seventy-five patients (59 males, 16 females), with an average age of 9 ± 2.1 years (range

6–12), were included. Table 1 provides demographic and diagnostic data. Fifty percent of

patients had one or more co-morbid DSM-IV Axis I conditions: disruptive behavior dis-

orders (20%), anxiety disorders (16%), autism spectrum disorders (12%), tic disorders

(5%), depressive disorders (3%).

Thirty-three patients were treated with atomoxetine (average dose at endpoint

1.2 ± 0.3 mg/kg) and 42 patients received stimulants (average endpoint dose in methyl-

phenidate equivalents 1.2 ± 0.4 mg/kg). There were no statistically significant differences

between atomoxetine or stimulant treated patients in age (p = 0.83), duration of follow-up

on a stable dose of medication (p = 0.91), initial GAF (p = 0.27), HALFS (p = 0.34),

CSI global score (p = 0.9), or initial 18 item ADHD rating scale score (p = 0.31).

At endpoint in the total sample, the 18 item ADHD rating scale improved with

7.6 ± 10.6 points (22% improvement), the CSI global score with 16.7 ± 27.6 points (23%

improvement), the GAF with 8 ± 7.9 points (16% improvement), and the HALFS with

1.8 ± 3.6 points (14% improvement). Seventy-nine percent of patients showed improve-

ment in their 18 item ADHD rating scale, while 21% worsened. The CSI global score

Table 1 Demographic and diagnostic data

Total sample (n = 75) Atomoxetine (n = 33) Stimulants (n = 42)

Age 9 ± 2.1 8.9 ± 2.3 9 ± 2

Duration of stable dose 9.9 ± 9.8 10 ± 11.2 9.8 ± 8.7

Initial GAF 50.9 ± 6.3 50 ± 6 51 ± 6.4

HALFS 13.2 ± 3 13.6 ± 2.8 12.9 ± 3.2

CSI global score 74.2 ± 27.5 73.7 ± 29.7 74.5 ± 26

18 item ADHD scale 34.5 ± 11.3 33 ± 11 35.7 ± 11.6

GAF Global Assessment of Functioning, HALFS Health and Life Functioning Scale, CSI Child SymptomInventory

Psychiatr Q (2011) 82:303–308 305

123

improved in 75% and worsened in 25% of subjects. The HALFS showed improvement in

61%, worsening in 24%, and no change in 15%. Similar percentages for the GAF were 83,

16, and 1%.

Table 2 shows the correlations between the change values of the different variables.

A multiple regression analysis with the improvement in the HALFS as dependent variable

and the 18 item ADHD rating scale change score and CSI change score as simultaneously

entered predictor variables showed a statistically significant result for the CSI change score,

but not for the 18 item ADHD rating scale change score (CSI: b = -0.062; p = 0.005; 18

item ADHD: b = -0.02; p = 0.75). The same result held with the GAF change score as

dependent variable (CSI: b = -0.11; p = 0.02; 18 item ADHD: b = -0.03; p = 0.82).

There were no statistically significant differences between atomoxetine (ATX) treated

patients and stimulant (STIM) treated patients in any of the outcome measure improve-

ments: 18 item ADHD scale (ATX -7.2 ± 10.1; STIM -7.9 ± 11; p = 0.78), CSI total

score (ATX -16.8 ± 29.2; STIM -16.6 ± 26.7; p = 0.97), HALFS (ATX 1.9 ± 2.7;

STIM 1.7 ± 4.2; p = 0.80) or GAF (ATX 9.5 ± 7.3; STIM 6.8 ± 8.3; p = 0.14).

Discussion

These community clinic patients with ADHD, treated with a stable dose of atomoxetine or

stimulants for an average length of time of approximately 10 months, showed a modest

improvement in symptom rating scales, global functioning and quality of life. The results

are very similar to two other studies from the same clinic with independent patient samples

[10, 14]. While the patients were significantly symptomatic at the start of treatment, as

manifested by an 18 item ADHD rating scale of 34.5/54, they continued to be quite

symptomatic with a score of 26.9/54 after 10 months on the same medication without any

adjustment in dose. The fact that no adjustments were made indicates, at some level, that

patient, parent and clinician were satisfied with the progress. On the other hand, an

improvement of 22% on the ADHD rating scale would not be considered a response in

many research studies. Many reasons can explain the difference between research and

community care: patient selection with specific exclusion criteria in research, minimal

level of follow-up in community clinics, and medical/psychiatric co-morbidity, among

others [8]. Why patients were maintained on the same dose of medication for a significant

amount of time with only modest improvement is unclear. It is possible that parent ratings

only capture improvements at home (mostly during the afternoon and evening) and not

during the day time in school, when medication effects may be more visible. Nevertheless,

parents may be satisfied with the treatment because of better school functioning.

Table 2 Correlations between improvements/change scores in variables

Variables Correlation R p value

CSI and 18 item ADHD 0.67 0.0001

CSI and GAF -0.43 0.0001

CSI and HALFS -0.38 0.0008

18 item ADHD and GAF -0.23 0.052

18 item ADHD and HALFS -0.26 0.02

CSI Child Symptom Inventory, GAF Global Assessment of Functioning, HALFS Health and LifeFunctioning Scale

306 Psychiatr Q (2011) 82:303–308

123

In regard to co-morbidity, it is interesting to note that the improvement in GAF and

quality of life were driven by the improvement in a global index of psychopathology (CSI),

not by the improvement in the ADHD rating scale. In adults, therapeutic improvement in

ADHD treatment with atomoxetine was reliably predicted by the presence of lifetime

co-morbid psychiatric disorders, especially depression [15]. If replicated in larger studies

with children, a need to broaden the treatment for patients with ADHD may arise: phar-

macotherapy for ADHD immediately associated with pharmacotherapy/psychotherapy for

co-occurring symptoms, not necessarily on a syndromal level, of disruptive behavior,

anxiety, depression, sleep, etc. In this regard, the treatment approach to ADHD may benefit

from a chronic disease management model, in which not only core symptomatology is

targeted, but consideration is given to co-morbidity, functional aspects, family pathology

and prevention of potential future impairments.

There were no differences between atomoxetine and stimulant induced improvements,

despite their different pharmacodynamic profile. For example, while both medications

have shown efficacy in treating core ADHD symptoms, atomoxetine may have more

effectiveness in anxiety and depression [16]. The fact that two pharmacodynamically

different treatments did not lead to different improvements in global psychopathology adds

to the concern that additional treatments may be necessary to get better results. Even

though, in adults, life time depression history predicted response to atomoxetine in ADHD

patients [15], it is of note that most patients in this sample did not have concurrent

significant elevation on anxiety and depression symptom scores.

Limitations of this study include a small sample size, naturalistic uncontrolled treat-

ment, and limited outcome measures. Strengths include the use of standardized and well-

validated assessment tools, and the use of specific outcome measures.

In summary, the improvement in quality of life during treatment for ADHD was driven by

an improvement in global psychopathology and not necessarily in the improvement of core

ADHD symptoms, indicating a need to broaden the treatment targets for ADHD patients.

Conflict of interest I have no potential conflict of interest pertaining to this Psychiatric Quarterlysubmission.

References

1. Barkley RA: Major life activity and health outcomes associated with ADHD. Journal of ClinicalPsychiatry 63(suppl 12):10–15, 2002

2. Matza LS, Paramore C, Prasad M: A review of the economic burden of ADHD. Cost Effectiveness andResource Allocation 3:5, 2005. http://www.resource-allocation.com/content/3/1/5

3. Gillberg C, Gillberg IC, Rasmussen P, et al.: Co-existing disorders in ADHD—Implications for diag-nosis and intervention. European Child & Adolescent Psychiatry 13(suppl 1):80–92, 2004

4. Stein REK: Measurement of ADHD outcomes: Implications for the future. Journal of Pediatric Psy-chology 32(6):728–731, 2007

5. Jensen PS, Martin D, Cantwell DP: Co-morbidity in ADHD: Implications for research, practice andDSM-V. Journal of American Academy of Child & Adolescent Psychiatry 36(8):1065-1-79, 1997

6. Spencer TJ: ADHD and co-morbidity in childhood. Journal of Clinical Psychiatry 67(suppl 8):27–31,2006

7. Jensen PS, Hinshaw SP, Kraemer HC, et al.: ADHD co-morbidity findings from the MTA study:Comparing co-morbid subgroups. Journal of American Academy of Child & Adolescent Psychiatry40(2):147–158, 2001

8. Jensen PS, Hinshaw SP, Swanson JM, et al.: Findings from the NIMH multimodal treatment study ofADHD (MTA): Implications and applications for primary care providers. Journal of Developmental &Behavioral Pediatrics 22(1):60–73, 2001

Psychiatr Q (2011) 82:303–308 307

123

9. Bastiaens L, Dello Stritto C: Quality of life of children and adolescents during psychiatric treatment in acommunity mental health setting. Psychiatry 2:32–35, 2005

10. Bastiaens L: Both atomoxetine and stimulants improve quality of life in an ADHD population treated ina community clinic. Psychiatric Quarterly 79:133–137, 2008

11. Sheehan DV, Lecrubier Y, Harnett Sheehan K, et al.: The validity of the Mini International Neuro-psychiatric Interview (MINI) according to the SCID-P and its reliability. European Psychiatry12:232–241, 1997

12. Gadow KD, Sprafkin J: Child Symptom Inventory-4. Stony Brook, NY, Checkmate Plus LTD, 199713. Bastiaens L, Dello Stritto C: Validity and reliability of the Health and Life Functioning Scale. Abstract

presented at the 51st annual meeting of the American Academy of Child and Adolescent Psychiatry,2004

14. Bastiaens L: Effectiveness and tolerability of atomoxetine in a real world ADHD population. Psychiatry4(12):44–48, 2007

15. Spencer TJ, Faraone SV, Michelson D, et al.: Atomoxetine and adult attention deficit hyperactivitydisorder: The effects of co-morbidity. Journal of Clinical Psychiatry 67:415–420, 2006

16. Kratochvil CJ, Newcorn JH, Arnold LE, et al.: Atomoxetine alone or combined with fluoxetine fortreating ADHD with co-morbid depressive and anxiety symptoms. Journal of American Academy ofChild & Adolescent Psychiatry 44(9):915–924, 2005

Author Biography

Leo Bastiaens, MD was born and raised in Belgium. After obtaining his MD degree from the University ofLeuven, Belgium in 1984, he graduated from his general psychiatry residency at Mount Sinai MedicalCenter, New York in 1988. He completed a child and adolescent psychiatry fellowship at HarvardUniversity, Boston in 1990. Dr. Bastiaens is board certified in general and child and adolescent psychiatry.He received training in cognitive therapy at the Cleveland Center for Cognitive Therapy, Ohio, and wascertified by the Academy of Cognitive Therapy in 2000. Dr. Bastiaens has practiced psychiatry in academictertiary care centers, psychiatric emergency rooms, community clinics, residential treatment facilities andaddiction programs. He has published over 30 peer-reviewed papers, book chapters, and researchpresentations on psychopharmacology, cognitive therapy, and addictions. He is a Clinical AssociateProfessor of Psychiatry at the University of Pittsburgh School of Medicine. He lives with his wife and twochildren in a suburb of Pittsburgh, PA. In his spare time, Dr. Bastiaens engages in long distance running. Hehas run marathons in Pittsburgh, Columbus, Cincinnati, Washington D.C., Erie, Harrisburg, Dayton, andBoston.

308 Psychiatr Q (2011) 82:303–308

123