Epi/ep,~iu,40(8):1 114-1121, 1999
Lippincott Williams & Wiikins, Inc., Philadelphia
0 international League Against Epilrpay
Clinical Research
Development of the Quality of Life in Epilepsy Inventory for
Adolescents: The QOLIE-AD-48
Joyce A. Cramer, "Lauren E. Westbrook, "Orrin Devinsky, *Kenneth Perrine,
?Marc B. Glassman, and $Carol Camfield
Yule University School of Medicine and VA Connecticut Healthcare System, Departments of Psychiatry and Neurology, New
Haven, Connecticut; *New York University School of Medicine and Hospital for Joint Diseuses, Department c?f Neurology,
Comprehensive Epilepsy Program, and fReseurch Analysis and Consultation, New, York, New York, U.S.A.; and jDalhousie
University School of Medicine and lzuak Walton Killam Gruce Health Center, Division of Neurology,
Halifax,Nova Scotia, Canada
Summery: Pur-pose: We report the development of an instrument to assess health-related quality of life (HRQOL) in adolescents with epilepsy.
Methods: A sample of 197 English-speaking adolescents
(aged 11-17 years) with epilepsy completed a test questionnaire of 88 items. Also included were mastery and self-esteem
scales to assess external validity. A parent simultaneously completed an I I-item questionnaire to evaluate the child's
HRQOL. Both adolescent and parent questionnaires were repeated in 2 4 weeks. Demographic information and information pertaining to seizures were collected at baseline along with
assessment of systemic and neurologic toxicity.
Results: The QOLIE-AD-48 contains 48 items in eight subscales: epilepsy impact (12 items), memorykoncentration (lo),
attitudes toward epilepsy (four), physical functioning (five),
stigma (six), social support (four), school behavior (four),
health perceptions (three), and a total summary score, with
higher scores indicating better HRQOL. Internal construct validity was demonstrated in a single-factor solution for the eight
Instruments have been developed to assess multiple
aspects of health-related quality of life (HRQOL) in
adults and children with epilepsy (1,2). Instruments to
assess HRQOL in children with epilepsy have used selfreport by the child and ratings by parents (3,4). The
Adolescent Psychosocial Seizure Inventory (APSI; 5) is
an adaptation of the Washington Psychosocial Seizure
Inventory (WPSI; 6 ) . There are parallels between the
APSI and the QOLIE-AD-48 in the inclusion of issues.
However, both the WPSI and APSI were designed to
Accepted January 6, 1999.
Address correspondence and reprint requests to J. A. Cramer at
Yale-VA Medical Center (G7E), 950 Campbell Avenue, West Haven,
CT 065 16-2770, U.S.A.; e-mail: [email protected].
dimensions. All correlations were statistically significant at p <
0.05 level. Internal consistency reliability estimated by Cronbach's alpha coefficient was 0.74 for the summary score and
ranged from a low of 0.52 (three-item Health Perceptions
Scale) to 0.73-0.94 for the other individual scales. Good testretest reliability was found for the overall measure (0.83). Summary score correlations with the two external validity scales,
self-efficacy and self-esteem were 0.65 and 0.54, respectively.
Statistically significant differences in summary scores indicating that HRQOL was increasingly better for adolescents as seizure
seventy decreases (no seizures = 77 ? 13, low = 70 17, high
= 63 f 17) were found among seizure-seventy groups.
Conclusions: These data describe the development of a robust instrument to evaluate HRQOL in adolescents with epilepsy. Empiric analyses provide strong evidence that the QOLIE-AD-48 is both a reliable and valid measure for adolescents with epilepsy. Key Words: Adolescents-QOLIEAD-48.
*
allow dichotomous (yes-no) responses that do not provide an opportunity for the respondent to express the
strength of conc-em for a problem. This design issue is
now considered important in instrument design (7). Two
scales have been developed recently for adolescents, but
not focused exchsively on epilepsy (89).
A full assessment of HRQOL issues in adolescents
(aged 11-17 years) is complex because of the wide range
of maturity within this age group, differences in independence and experience, and potential volatility of emotions (8-10). To consider adolescents as youllg adults is
problematic because of these developmental issues. Simply changing the language in a questionnaire designed
for adults is inappropriate (e.g., change work to school,
leisure activities to play activities) because adolescents
1114
1115
EVALUATING HRQOL IN ADOLESCENTS WITH EPILEPSY
not only have different activities and levels of responsibility, but they also have different interests and areas of
concern from those of adults. Thus, development of an
HRQOL instrument for adolescents with epilepsy was
deemed necessary and required a thorough review of the
domains, as well as a range of items pertinent to teenagers. Having developed the QOLIE-89 (1 l), QOLIE-3 1
(12), and QOLIE-10 (13) for adults with epilepsy, we
had to consider new constructs from the perspective of
adolescents before developing a test questionnaire for
this group. We report the initial development and psychometric properties of an instrument that focuses on
specific areas of HRQOL for adolescents with epilepsy:
the QOLIE-AD-48.
METHODS
Subjects
One hundred ninety-seven adolescents (aged 1 1-17
years) with epilepsy were enrolled at 22 sites (17 epilepsy centers in the United States, two in Canada, and
three private practices) after fully informed consent (approved by the Institutional Review Boards at each site).
Eligibility criteria included age, 11-17 years; ability to
read English at the fifth-grade level (determined by the
Wechsler Reading Achievement Test or similar); active
epilepsy within the past 2 years; duration of epilepsy, >1
year; and availability of a parendguardian able to speak
and read English. Subjects were excluded if they had
brain surgery in the past year, a change in antiepileptic
drug (AED) regimen in the past 4 months, used a concomitant medication with central nervous system (CNS)
effects, or had another progressive neurologic or psychiatric illness. Patients with a diagnosis of primary generalized (absence, myoclonic, tonic-clonic seizures) or
partial (simple partial, complex partial, secondarily generalized tonic-clonic seizures) epilepsy were included.
Table I describes the categorization of seizure severity
based on the type and number of seizures experienced in
the past 12 months.
Measures
HRQOL test questionnaire for adolescents
To assess the quality of life of adolescents with epilepsy, a self-report, self-administered test instrument was
developed under the guidance of a panel of seven experts
in the field (QOLIE-AD Advisory Group). Domains and
specific items were derived from several sources. Some
items were adapted from generic instruments: overall
QOL item from a patient preferences study (14) and from
the Faces Scale (I 5 ) . Other items were selected from the
RAND 36-item Health Survey 1.O (a.k.a. SF-36) (16,17)
and were adapted for adolescents by changing the phrase
to reflect activities for this age group. From the SF-36,
four of five items from the emotional well-being sub-
TABLE 1. Seizure severity categories: seizure frequency in
the past 12 months"
Seizure type
Simple partial (n = 25)'
Complex partial (n = 70)'
Generalized tonic-clonic
(n = 61)'
Absence (n = 30)'
Myoclonic (n = 13)'
No
seizures'
(n = 41)
Low seizure
severity
(n = 52)
High seizure
severity
(n = 104)
0
0
1-20
I 4
>20
0
0
1
1-20
1-20
>I
>20
>20
0
>4
~
'' Patients with multiple seizure types were classified according to the
most severe category of all seizure types.
All patients in the seizure-free group had at least one seizure in the
12- to 24-month period preceding enrollment.
' Number of subjects experiencing this type of seizure (some suhjects had more than one seizure type).
scale, four items each from the physical function, rolephysical, and role-emotional subscales, and two items
from the vitality subscale were included in the test questionnaire. Epilepsy-targeted items were related to issues
commonly reported by adolescents with epilepsy. These
questions were derived from focus groups at the Baltimore, Halifax, Miami, and New York City sites, supplemented by experiences of health professionals who
worked with this population, and a literature review of
psychosocial aspects of epilepsy for adolescents (5,18,
19). The focus groups (five to 10 adolescent inpatients
and outpatients) identified topics of interest related to
epilepsy and commented on topics raised by the moderatortclinician. Specific domains expected to be pertinent to these groups were raised for discussion (e.g.,
based on clinical experience and literature review). New
items were written, or existing questions were modified
to reflect the issues and circumstances particular to adolescents.
Initially, items were allocated to domains on conceptual grounds. In addition, the Child Attitude Toward 111ness Scale (CATIS; 4) was included in its entirety; the
Personal View Section of The Perceived Stigma of Epilepsy: Adolescent Form, currently under development,
also was selected based on previous research (L. Westbrook, unpublished observations). A total of 88 QOL
items was included in the test questionnaire. In addition,
a single open-ended question asked subjects to comment
on the questionnaire and note topics that were not covered.
The questionnaire was divided into two sections as ( I )
generic (physical, mental, emotional), and (2) epilepsy
specific (the effects of epilepsy and antiepilepsy medication. Section 1 topics included overall health perceptions, physical activities, impact of physical or emotional
problems, mood, concentration, memory, cognitive
skills, and social support from family and friends. Section 2 topics included impact of epilepsy and AEDs on
activities, fears about seizures, perceived severity of seiEpilrpsiu, Vol. 40, No. 8, 1999
1116
J . A. CRAMER ET AL.
zures, adverse mental and physical effects of medication,
and role limitations, including driving. Items from the
stigma scale and CATIS (4) were included in this section.
External-validity measures
Two measures were used to assess external construct
validity: the Rosenberg Self-Esteem Scale (RSE; 20) and
the Ilfeld Self-Efficacy Scale (ISE; 21). The RSE is a
1 0-item scale that is widely used to assess global feelings
of self-acceptance and self-worth. Reliability (Cronbach’s alpha, 0.78) and validity are well documented.
The ISE is a well-established, seven-item tool designed
to tap perceived degree of mastery over various life situations (Cronbach’s alpha, 0.86). The concepts tapped by
these scales were expected to correlate with adolescent’s
quality of life. These domains were selected based on
previous research (22).
HRQOL questionnaire for parents of
adolescent subjects
Parents provided an additional (proxy) source of external information with a self-report, self-administered
questionnaire. The 11-item parent questionnaire evaluated the child’s quality of life in five areas: general perceived health, impact of physical and emotional problems on activities, limitations caused by epilepsy or
medications on activities, energy level, and concentration and memory problems. Parents were instructed to
answer “the way you believe best describes your child’s
situation.”
Neurotoxicity and systemic toxicity scales
A directed neurologic examination and interview focusing on signs and symptoms of neurotoxicity was conducted by a neurologist by using a modification of the
VA Neurotoxicity Rating Scale (23). The presence and
severity of problems with the following items were
noted: diplopia, nystagmus, dysarthria, gait, rapid alternating movements, tremor, sedation, affect, mood, cognitive function, dizziness or lightheadedness, headaches,
and other drug-related neurotoxic effects. A modification
of the VA Systemic Toxicity Rating Scale (23) assessed
the following signs and symptoms related to drug use:
gastrointestinal problems, weight gain, changes in hair
quality or quantity, and other drug-related systemic effects. Individual item scores can range from 0 to 50
points on both scales.
Clinical, demographic, and health-care
utilization variables
Patients and parents were interviewed by clinical staff
to determine seizure type and frequency during the past
year, number and types of current AEDs, age, gender,
school level (regular or special classes), current school
grades for math and English, and primary language spoken at home. Health-care utilization in the past year in-
cluded number of doctor visits, nights in hospital, as well
as concurrent medical and neurologic conditions. Patients and parents also were interviewed to ascertain socioeconomic status (educational achievement, occupational status for both parents), grade level, class type
(special, regular, mixed), current school grade, level of
participation in extracurricular and social activities
(number of hours spent per month in activities outside of
school, and number of close friends), and family history
of epilepsy (presence in either parent).
Data collection
Eligible adolescents with epilepsy and their parent
who agreed to participate in the study after full informed
consent were interviewed to obtain demographic and
personal information. The patient was examined for neurologic and systemic toxicity. Thereafter, both patient
and parent completed the HRQOL questionnaires. Time
2, administration of the Adolescent and Parent Questionnaires, occurred 2 4 weeks later. The majority were
completed at home and returned by mail. Time 2 analysis
did not include retest questionnaires completed >4 weeks
after Time 1 or data from subjects reporting a major
stressful life event during the intervening weeks. Complete data for Time 2 were available on 85% of the
sample. Participants received $45 per family from the
coordinating research site as compensation for their time
and travel.
Analysis plan
Data were entered and edited at the coordinating center before being sent to an independent statistical consultant (M.G.) for analyses. Factor analysis by using a
double cross-validation procedure was conducted on the
adolescent quality of life test questionnaire to define operationally the factors and items in the final instrument.
A summary score was then derived from the resulting
factors. Scale scores were first recalibrated to 0-100
points. The standardized (beta) partial-regression coefficients for each of the factors were summed, and each
regression coefficient was then divided by that sum to
derive relative weights. Each factor was then multiplied
by its relative weight, and the sum of these products
resulted in an overall summary (total) score.
Internal consistency reliability was estimated for each
of the final factor-based scales as well as for the overall
summary score with Cronbach’s alpha coefficient. Pearson’s product-moment correlations and intraclass correlation coefficients determined test-retest reliability of the
summary score. Correlations also were calculated for the
summary score with the two external-validity scales.
Subscale scores were transformed linearly into scales of
1-100 points, with higher values representing better
functioning and well-being. Descriptive statistics
(means, SDs ranges, minimudmaximum scores) were
calculated for each scale and the summary score.
EVALUATING HRQOL IN ADOLESCENTS WITH EPILEPSY
1117
We hypothesized that adolescents having fewer and
less severe seizures (i.e., no seizures, low-seizureseverity groups) would have better HRQOL than those
with more frequent and severe seizures (i.e., highseizure-severity group). We also expected positive correlation with the other external-validity scales such that
higher self-esteem and higher self-efficacy correlate with
better HRQOL.
were retained. Thus the eight scales and 48 items that
constitute the final questionnaire were found in both
half-samples and the whole sample. Table 3 lists the
eight subscales, number of items, mean scores, and standard deviations for each. The range of scores for all
subscales was 0-100 (indicating a good range of responses), with means ranging from 39.8 -t 22.8 to 90.3 &
15.4. The complete instrument and scoring instructions
are presented in the Appendix.
RESULTS
Content validity
Content validity is the extent to which the instrument
measures a representative range of the attributes under
study. Items were selected based on a literature review,
existing measures, focus groups with adolescents with
epilepsy, and the experience of a panel of experts. The
appropriateness of item selection (described in Methods)
was further supported by asking subjects to comment
about additional HRQOL concerns not covered in the
questionnaire. Only 19% of subjects made any additional
comments for a total of 38 responses. A qualitative
analysis showed that most of the comments were questions about issues unrelated to or already included in the
questionnaire. Several subjects described personal feelings about having epilepsy. None of the comments revealed any other topics not covered in the questionnaire.
Description of the sample
The sample consisted of 101 girls and 96 boys with a
mean age of 14.2 ? 1.74 years (range, 11-17 years) and
mean age of seizure onset of 7.5 ? 4.2 years, and mean
duration of epilepsy of 6.6 2 4.2 years. Additional sociodemographic data are listed in Table 2.
First-order factor analysis: Double
cross-validationprocedure
An empirically rigorous approach was taken for questionnaire development by using a double cross-validation
(otherwise known as split-sample) procedure. The data
from the 197 patients were divided into two random
half-samples. A list-wise deletion method was used so
that if a case was missing any single item in a factor or
subscale, the case was deleted from these analyses. That
resulted in a sample of 191 complete cases. The 88 items
were submitted to the first-order factor analysis. Factors
were generated in the first half-sample and then applied
to the second half-sample to determine whether the same
factor structure would prevail in both samples. The step
was repeated, starting with factors generated from the
second half-sample that were then applied to the first
half-sample. The final factor solution used the entire
sample and factors. Items that both matched in the two
half-samples and were replicated in the whole sample
TABLE 2. Sociodemographic characteristics of the
patient population
Gender
Age (Yr)
Age at onset (yr)
Duration of epilepsy (yr)
Number AEDs
School grade
Class level
Ever repeat school grade
School performance
No. siblings
Home language
Parents’ education
Parents’ work
Parent history of epilepsy
AED, antiepileptic drug.
51% female (n, 101), 49% male (n, 96)
14.2 f 1.74; range, 11-17
7.5 r 4.3; range, 0-15
6.6 k 4.2; range, 0-17
1.44 5 0.7; range, 0 4
8.7 5 1.8; range, 5-12
75% regular, 5% special, 20% mixed
20% (n, 39) held back one grade
(grade held back range, 1-7)
21 of 39 (54%) held back first grade
86% math grade, 92% English
grade = A, B, C , level
0, 13%; I , 37%; 2, 27%; 2 3 , 23%
96% English as primary language
50% mothers, 47% fathers > high school
68% mothers, 83% fathers
work part- or full-time
10% mother, 4% father
Construct validity
Construct validity is the evidence that a construct is
consistent with its referent theoretic properties. We established the construct validity of the QOLIE-AD-48 in
several ways. First, we examined the internal structure of
the questionnaire to determine whether we could support
empirically that our multidimensional measure could reflect a single construct. A principal-axis factor analysis
of the eight dimensions was conducted by using an oblique (Promax) rotation. Two second-order factors were
suggested. However, the second factor was operationally
defined by only two subscales with low rotated pattern
loadings (social support, 0.42, and school behavior,
0.41). This suggested that the QOLIE-AD-48 items and
eight subscales could be subsumed under a single construct that we labeled HRQOL. This analysis also provided justification for the development of an overall
summary score of adolescent HRQOL. The contribution
of each of the subscales to the summary measure can be
viewed as a structure coefficient (Pearson correlations)
between the second-order factor and each of the eight
first-order factors (subscales) from which it is composed
(Table 3, column 8). All but the social-support subscale
demonstrated moderate to strong positive correlations
with the summary measure. Nonetheless, this dimension
was retained for its contribution on theoretic grounds.
Construct validity also was evaluated by measuring
correlations between the summary score and other scales
(i.e., expecting a relation between the external measure
Epilepsia, Vol. 40, No. 8, 1999
1118
J. A. CRAMER ET AL.
TABLE 3. Descriptive statistics and reliuhilities for QOWE-AD-48 subscales“
Suhscale
No.
items
Mean
score”
SD
Cronbach’s
alpha
coefficient
Epilepsy Impact
Memory-Concentration
Attitudes
Physical Function
Stigma
Social Support
School Behavior
Health Perceptions
Total score
12
10
4
5
6
4
4
3
48
70.6
67.6
39.8
63.6
71.3
72.4
90.3
65.8
67.7
26.9
22.4
22.8
30.6
22.0
24.4
15.4
19.1
17.3
0.94
0.9 I
0.83
0.86
0.75
0.80
0.73
0.52
0.74
~
”
’I
~
~
~
~
~
Minimum
score
Maximum
score
0
I00
100
100
0
0
0
0
0
0
0
I6
~
Structure Coefficient:
Overall epilepsy
experience
100
100
100
100
I00
98
0.93
0.72
0.57
0.52
0.64
0.08
0.37
0.66
-
~~
Data for six subjects who experienced important life events between test administrations were deleted.
All aubscale scores were linearly transformed to range from 0 to 100 points, with higher values representing better functioning.
that assesses the theory underlying the HRQOL assessment and the test instrument). The self-esteem and selfefficacy scales were hypothesized to correlate with the
HRQOL constructs. The overall summary scale moderately to strongly correlated with both scales ( r = 0.65,
self-esteem; 0.54, self-efficacy). In addition, the parental
questionnaire correlated well with the summary score
(0.67), indicating similar perceptions between parents
and adolescents.
Reliability
Reliability data are presented in Table 3. Internalconsistency reliability, assessed with Cronbach’s alpha
coefficient, was satisfactory, with all subscales above the
conventional standard of 20.70 (range, 0.73-0.94) except for health perceptions (0.52). No floor or ceiling
effects were seen. Test-retest reliability was 0.83 for the
summary (total) score. The sensitivity of the QOLIEAD-48 to differences among groups was evaluated by
examining the relation of the summary score to the seizure-severity categories. QOLIE-AD-48 scores significantly decreased (worsened) as seizure severity increased (Table 4).
DISCUSSION
These data describe the QOLIE-AD-48 as a comprehensive and sensitive instrument for the assessment of
HRQOL among adolescents with epilepsy. We established the psychometric characteristics of the QOLIEAD-48, documenting that the instrument has content and
construct validity, internal consistency, reliability, and
test-retest reliability. The questionnaire can be comTABLE 4. QOLIE-AD-48 surnrnaqv scores b j seizure
severity category
Mean
?
SD
No-seizure
group
Low-severity
group
77.3 k 12.6
70.3 f 16.6
~
High-severity
group
62.7 k 16.5
~~~~
Two-tailed t tests comparing groups: No versus low; p = 0.04; No
versus high: p = 0.001; Low versus high; p = 0.006.
EpiI?psia, Vol. 40. N o . 8, 1999
pleted in 15-20 min and provides information about a
variety of issues pertinent to this age group.
No reliable generic HRQOL instrument for this age
group was available at the time the QOLIE-AD-48 was
designed. Since that time, two instruments have been
reported. The Quality of Life Profile: Adolescent Version (QOLPAV) (8) is a 54-item instrument covering
three broad domains (being, belonging, and becoming).
Adolescents are asked to rate items for importance and
satisfaction. The “1 6D” measure (9) includes 16 dimensions. Initial assessment included normal schoolchildren
aged 12-15 years compared with small groups of adolescents with other medical disorders including 32 with
epilepsy. Further evaluation of these instruments in adolescents with epilepsy is needed. In addition, other data
collected during this study will be published elsewhere
(e.g., comparison between adolescent and parental perceptions, correlation of QOLIE-AD-48 scores and adverse effects, health-care use).
Thirty-eight adolescents wrote comments at the end of
the questionnaire. None of the comments suggested areas
of concern that had been omitted in the questionnaire.
None of the comments showed negative feelings about
completing the questionnaire (e.g., items or time), indicating the acceptability of a HRQOL questionnaire by
adolescents. The most poignant note was “This questionnaire made me realize some feelings that I just don’t feel
comfortable sharing with my family.” We take this note
as an example of how HRQOL assessment might open
an awareness of and communication about HRQOL
among adolescents, families, and clinicians.
During the careful approach to development of this
instrument, only those items that met predefined and rigorous statistical standards were retained. Several items
that were empirically expected to be important did not
meet minimal statistical standards for retention and were
thus dropped from the QOLIE-AD-48. Nonetheless, we
recognize that several of the items (seizure worry, fear of
injury, and fear of dying) may be extremely important in
evaluating individual cases. These three items may be
1119
EVALUATING HRQOL IN ADOLESCENTS WITH EPILEPSY
appended to the QOLIE-AD-48 at the investigator’s discretion, with the caveat that they are not included in any
subscale or the total score (see Appendix).
As with all instrument-validation studies, these results
represent only the test population, albeit a geographically
and medically varied group. As such, the values presented
are not population norms. The key test for this instrument is
whether it detects change among intervention groups.
Based on the empiric analyses described, the QOLIEAD-48 is robust measure, with good psychometric characteristics, for the evaluation of HRQOL in adolescents
with epilepsy. Because of its simplicity, the summary
score should be a convenient way to express HRQOL.
The combination of epilepsy-related issues with generic
HRQOL issues provides a broad basis for understanding
how adolescents integrate the chronic seizure disorder
into daily life. Availability of this instrument might facilitate comparisons with adolescents who have other
chronic disorders, such as asthma, migraine, arthritis, or
diabetes. As with all newly developed instruments, its
strengths and weaknesses will be identified through use.
Future studies will determine its usefulness in clinical
trials of AEDs, surgical, psychosocial, and other interventions, as well as in longitudinal studies.
The following sites and individuals participated in this
project: Hospital for Joint Diseases, New York, NY (0.
Devinsky, H. Abramson); IWK Grace Health Center,
Halifax, NS (C. Camfield); Children’s Hospital, Vancouver, BC, Canada (K. Farrell, M. Berg); University of
Kentucky, Lexington, KY (S. Bauman, M. Ryan); Rutgers Medical Center, New Brunswick, NJ (H. Mandelbaum); Montefiore Medical Center, Bronx, NY (S.
Shinnar, C. O’Dell); Henry Ford Hospital, Detroit, MI (G.
Barkley, B. Rader); Children’s Hospital Medical Center,
Cincinnati, OH (M. Bare); Children’s Hospital Center,
Washington, DC (S. Weinstein, M. Kolodgie); Nemours
Children’s Clinic, Jacksonville, FL (W. Turk, L. Bailet);
Epi-Care Center, Nashville, TN (G. Montouris, S. Lancaster); Southern Illinois University, Springfield, IL (A.
Morales, D. Wildrick); Yale Medical Center, New Haven, CT (R. Novotny, C. Cardoza); Miami Children’s
Hospital, Miami, FL (M. Duchowny, P. Dean); Stanford
Medical Center, Stanford, CA (M. Morrell, K. Seal);
University of North Carolina, Chapel Hill, NC (S. Tennison, s. Lannon); Long Island College Hospital, Lamm
Institute, Brooklyn, NY (H. Bennett); University of
Texas Medical Branch, Galveston, TX (J. Hartshorn, D.
Ray); Neptune, NJ (Pietrich); Children’s Medical Center,
Dallas, TX (D. Riela, B. Singh, D. Calligaro-Wharten);
University of Wisconsin, Madison, WI (B. Gidal); and
Johns Hopkins Medical Center, Baltimore, MD (E. Vining, D. Pillas).
Acknowledgment: J. Austin, P. Dean, E. Vining, and D.
Pillas provided assistance in the conceptualization of a ques-
tionnaire for adolescents. T h e QOLIE Development Group appreciates donations from Abbott, Astra-Chamwood, HoechstMarion Roussel, Novartis, Parke-Davis, and others toward development of t h e QOLIE-AD-48. T h e QOLIE-AD-48 is
copyright by the QOLIE Development Group, 1998. It is available at no charge. Contact [email protected].
REFERENCES
1. Cramer JA. Quality of life assessments in epilepsy, In: Spilker B,
ed. Quality of life and pharmacoeconomics in clinical trials. Phila-
delphia: Lippincott-Raven, 1996909-18.
2. Carpay HA, Arts WFM. Outcome assessment in epilepsy: available rating scales for adults and methodological issues pertaining
to the development of scales for childhood epilepsy. Epilepsy Res
1996:24: 127-36.
3. Hoare P, Russell M. The quality of life of children with chronic
epilepsy and their families; preliminary findings with a new assessment measure. Dev Med Child Neurol 1995;37:689-96.
4. Austin JK, Huberty TJ. Development of the Child Attitude Toward
Illness scale. J Pediatr Psychol 1993; 18:467-80.
5. Batzel LW, Dodrill CB, Dubinsky BL, et al. An objective method
for the assessment of psychosocial problems in adolescents with
epilepsy. Epilepsia 1991;32:202-1 I .
6 . Dodrill CB, Batzel LW, Quiesser HR, Temkin N. An objective
method for the assessment of psychological and social problems
among epileptics. Epilepsia 1980;21: 123-35.
I . Cramer JA, Spilker B. Quulity of life and pharmacoeconomics: un
introduction. Philadelphia: Lippincott-Raven, 1 998.
8. Raphael D, Rukholm E, Brown I, Hill-Bailey P, Donato E. The
quality of life profile-adolescent version: background, description,
and initial validation. J Adolesc Health 1996; 19:366-75.
9. Apajasdlo M, Sintonen H, Holmberg C, et al. Quality of life in
early adolescence: a sixteen-dimensional health-related measure
(16D). Qua1 Life Rer 1996;5:205-11.
10. Freeman JM. Vining EPG, Pillas DJ. Seizures and epilepsy in
childhood: a guide for parents. Baltimore: The Johns Hopkins
University Press, 1990.
11, Devinsky 0, Vickrey BG, Perrine K, et al. Development of an
instrument of health-related quality of life for people with epilepsy.
Epilepsia 1995:36:1089-104.
12. Cramer JA, Perrine K, Devinsky 0, Bryant-Comstock L, Meador K,
Hermann BP. Development and cross-culturaltranslation of a 3 1-item
quality of life questionnaire (QOLIE-31). Epilepsia 1998;39:81-8.
13. Cramer JA, Perrine K, Devinsky 0, Meador K. A brief questionnaire to screen for quality of life in epilepsy: the QOLIE-10. Epilepsia 1996;37:577-82.
14. Hadom D, Hays RD. Multitrait-multimethod analysis of healthrelated quality of life preferences. Med Care 1991;29:829-40.
15. Andrews FM, Withey SB. Social indicators of well-being: Americans’ perception of life quality. New York: Plenum Press, 1976.
16. Ware JE, Sherbourne CD. A 36-item Short Form Health Survey
(SF-36): 1. conceptual framework and item selection. Med Care
1992;30:473-83.
17. Hays RD, Sherbourne CD, Maze1 E. The RAND 36-item health
survey I .O. Health Econ 1993;2:212-27.
18. Mittan RJ. Fear of seizures. In: Whitman S, Hermann BP, eds.
Psychopathology in epilepsy: social dimensions. New York: Oxjord University Press, 1986:90-1 2 I.
19. Hermann BP. The relevance of social factors to adjustment in
epilepsy. In: Devinsky 0, Theodore WH, eds. Epilepsy and behavior. New York: Wiley-Liss, 1991:23-36.
20. Rosenberg M. Society and the adolescent selj-image. Princeton,
NJ: Princeton University Press, 1965.
21. Ilfeld FW. Psychological status of community residents along major dimensions. Arch Gen Psychiatiy 1978;35:716-24.
22. Jacoby A. Epilepsy and the quality of everyday life: findings from
a study of people with well-controlled epilepsy. Soc Sci Med 1992;
34:657-66.
23. Cramer JA, Smith DB, Mattson RH, Delgado-Escueta AV, Collins
JF, and the VA Epilepsy Cooperative Study Group. A method of
quantification for the evaluation of antiepileptic drug therapy. Neurology 1983;33(suppl 1 ):26-37.
Epilepsia. Vol. 40, No. 8, I999
1120
J. A. CRAMER ET AL.
APPENDIX
Quality of life in epilepsy for adolescents: QOLIE-AD-48, Patient Inventory, Version 1.0, U S . English
PART 1: GENERAL HEALTH
I . In general, would you say your health is: Excellent = 5 , Very good = 4, Good = 3, Fair = 2, Poor = 1
2. Compared to 1 year ago, how would you rate your health in general now? Much better = 5 , Somewhat better
same = 3, Somewhat worse = 2, Much worse = 1
3.
4.
5.
6.
7.
8.
9.
10.
11.
=
4, About the
The following questions are about activities you might do during a TYPICAL DAY. We want you to answer how much your
health limits you in these activities. Very often = 1, Often = 2, Sometimes = 3, Not often = 4, Never = 5.
In the past 4 weeks, how often has your health limited:
Heavy activities, such as running, participating in very active sports (such as gymnastics, roller-blading, skiing)?
Moderate activities (such as walking to school, bicycle riding)?
Light activities (such as carrying packages or a school bag full of books)?
Other daily activities (such as taking a batwshower alone, going to and from school alone)?
The following questions are about your regular daily activities, such as chores at home, babysitting, attending school, being
with friends and family, doing homework, or taking part in afterschool activities and lessons. We want to know if you had any
of the following difficulties with your regular activities as a result of any physical problems (such as illness) or emotional
problems (such as feeling sad or nervous)? Very often = 1, Often = 2, Sometimes = 3, Not often = 4, Never = 5.
In the past 4 weeks, how often have physical or emotional problems caused you to:
Limit the kind of schoolwork, chores, sports, or other activities you did?
Have difficulty performing the schoolwork, chores, sports, or other activities you did (for example, it took extra effort)?
In the past 4 weeks, how often:
Did you skip school for no reason?
Were you in trouble in school (with teachers or other stafo?
Were you in trouble out of school (with police, security guards, bus driver, etc.)?
These questions are about how you FEEL and how things have been for you during the past 4 weeks. For each question, please
indicate the one answer that comes closest to the way you have been feeling. (All of the Time = 1, Most of the time = 2,
Some of the time = 3, A little of the time = 4, None of the time = 5 ) .
In the past 4 weeks, how often have you:
12. Had trouble concentrating on an activity?
13. Had trouble concentrating on reading?
14.
15.
16.
17.
18.
19.
20.
The following questions are about mental activities and language problems that may interfere with your normal schoolwork
or living activities. (All of the time = 1, Most of the time = 2, Some of the time = 3, A little of the time = 4, None of the
time = 5 ) .
In the past 4 weeks, how often have you:
Had difficulty thinking?
Had difficulty figuring out and solving problems (such as making plans, making decisions, learning new things)?
Had a problem with complicated projects that require organization or planning like computer games or difficult homework)?
Had trouble remembering things you read hours or days before?
Had trouble finding the correct word?
Had trouble understanding your teachers?
Had trouble understanding what you read?
21.
22.
23.
24.
The following questions ask about the support you get from others (including family and friends) (Very often = 5, Often =
4, Sometimes = 3, Not often = 2, Never = 1).
Have someone available to help you if you needed and wanted help?
Have someone you could confide in or talk to about things that were troubling you?
Have someone you could talk to when you were confused and needed to sort things out?
Have someone who accepted you as you were, both your good points and bad points?
PART 2: EFFECTS OF EPILEPSY AND ANTIEPILEPSY MEDICATIONS
The following questions ask about how your epilepsy or medications (antiepileptic drugs) have affected your life in the past
4 weeks. (Very often = I , Often = 2, Sometimes = 3, Not often = 4, Never = 5).
In the past 4 weeks, how often did you:
25. Feel that epilepsy or medications limited your social activities (such as hanging out with friends, doing extracurricular activities)
compared with social activities of others your age‘?
26. Feel alone and isolated from others because of your epilepsyheizures?
27. Miss classes because of seizures or medications?
28. Use epilepsy or medication side effects as an excuse to avoid doing something you didn’t really want to do?
29. Feel embarrassed or “different” because you had to take medications?
30. Feel that epilepsy or medications limited your school performance?
3 I . Feel you had limitations because of your seizures?
Epilqi,siu, Vol. 40, No. X , 1999
1121
EVALUATING HRQOL IN ADOLESCENTS WITH EPILEPSY
32. Feel that epilepsy or medications limited your independence?
33. Feel that epilepsy or medications limited your social life or dating?
34. Feel that epilepsy or medications limited your participation in sports or physical activities?
The following question asks about possible side effects from antiepileptic drugs.
In the past 4 weeks, how did you feel (Very bad = 1, Bad = 2, O.K. = 3, Good = 4, Very good = 5):
35. About how you looked (side effects such as weight gain, acne/pimples, hair change, etc.)?
In the past 4 weeks, how much were you bothered by (A lot = 1, Some = 2, Not much = 3, A little = 4, Not at all
36. Limits set by parentslfamily because of your epilepsy or medications?
37.
38.
39.
40.
41.
42.
=
5):
Next are some statements people with epilepsy sometimes make about themselves.
For each statement, circle the answer that comes closest to the way you have felt about yourself in the past 4 weeks (Strongly
agree = 1, Agree = 2, Disagree = 3, Strongly disagree = 4):
1 consider myself to be less than perfect because I have epilepsy.
If I applied for a job, and someone else also applied who didn’t have epilepsy, the employer should hire the other person.
I can understand why someone wouldn’t want to date me because I have epilepsy.
I don’t blame people for being afraid of me because I have epilepsy.
I don’t blame people for taking my opinions less seriously than they would if I didn’t have epilepsy.
I feel that my epilepsy makes me mentally unstable.
The following questions ask about your attitudes toward epilepsy. Circle one number for how often in the past 4 weeks you have
had these attitudes. [Very (negative) = 1 , A little (negative) = 2, Not sure = 3, A little (positive) = 4, Very (positive) = 51.
43. How good or bad has it been that you have epilepsy?
44. How fair has it been that you have epilepsy?
45. How happy or sad has it been for you to have epilepsy?
46. How bad or good have you felt it is to have epilepsy?
47. Feel that your epilepsy kept you from doing things you like to do?
48. How often do you feel that your epilepsy kept you from starting new things (Very often = I , Often = 2, Sometimes = 3, Not
often = 4, Never = 5)?
Optional extra items:
In the past 4 weeks, how often have you felt (Very often = I , Often = 2, Sometimes
Worry about having another seizure?
Fear dying because of seizures?
Worry about hurting yourself during a seizure?
QOLIE-AD-48: Item allocation to domains:
1. Epilepsy impact (12 items): 25, 26,27, 29, 30, 31, 32,
33, 34, 36, 47, 48.
2. Memorykoncentration (10 items): 8, 12, 13, 14, 15,
16, 17, 18, 19, 20.
3. Attitudes toward epilepsy (4 items): 43, 44, 4.5, 46.
4. Physical functioning ( 5 items): 3, 4, 5 , 6, 7.
5. Stigma (6 items): 37, 38, 39, 40, 41, 42.
6. Social support (4 items): 21, 22, 23, 24.
7. School behavior (4 items): 9, 10, 11, 27.
8. Health perceptions (3 items): 1, 2, 35.
Scoring instructions:
1. Convert the precoded numeric values of items to
0-100 point scores, with higher converted scores always reflecting better quality of life. For example,
items with a four-point response are coded as 0, 33.3,
66.7, 100; items with a five-point response are coded
as 0, 25, SO, 75, 100.
2. Calculate the mean value of the items in each subscale. Adjust the denominator to include only items
answered.
3. To calculate the total score, multiply the mean by the
relative weight for the subscale:
=
3, Not often
=
4, Never = 5 )
Epilepsy impact
= 0.31
Memorykoncentration
= 0.17
Attitudes
= 0.09
Physical function
= 0.09
Stigma
= 0.13
Social support
= 0.02
School behavior
= 0.06
Health perceptions
= 0.12
4. Add the weighted values for all subscales to determine the total (summary) score.
5. T-scores can be determined for each subscale and the
total (summary) score to represent linear transformations with a mean of 50 and standard deviation of 10
for the cohort evaluated. Higher T-scores reflect a
more favorable quality of life. The formula for computing the T-score is as follows:
r
T-score = 50 +
I
L
/ Observed final scale score \ 1
\minus scale mean in Table 3 )
10
( Scale standard 1
\deviation in Table 3)
_I
I
QOLIE-AD-48, copyright by the QOLE Development
Group. Permission for use will be granted at no cost on
written request. A single copy of the formatted questionnaire and scoring sheet will be provided with permission.
Epilepsia, Vol. 40, No. 8, 1999