Epilepsy & Behavior 37 (2014) 26–31
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Developing the Pediatric Refractory Epilepsy Questionnaire: A pilot study Vaishnavi Purusothaman a,⁎, Robin C.C. Ryther b, Mary Bertrand b, Lisa A. Harker c, Donna B. Jeffe d, Michael Wallendorf e, Matthew D. Smyth f, David D. Limbrick f a
Division of Biology and Biomedical Sciences, Washington University in St. Louis Graduate School of Arts and Sciences, St. Louis, MO, USA Department of Neurology, Saint Louis Children's Hospital, Washington University in St. Louis School of Medicine, St. Louis, MO, USA Department of Psychology, Saint Louis Children's Hospital, Washington University in St. Louis School of Medicine in St. Louis, St. Louis, MO, USA d Division of General Medical Sciences, Department of Medicine, Washington University in St. Louis, School of Medicine St. Louis, MO, USA e Division of Biostatistics, Washington University in St. Louis School of Medicine in St. Louis, St. Louis, MO, USA f Department of Neurological Surgery, Saint Louis Children's Hospital, Washington University in St. Louis, School of Medicine St. Louis, MO, USA b c
a r t i c l e
i n f o
Article history: Received 5 January 2014 Revised 25 April 2014 Accepted 29 April 2014 Available online xxxx Keywords: Medically refractory Intractable Seizure Epilepsy Questionnaire Epilepsy surgery Palliative epilepsy surgery
a b s t r a c t Purpose: Up to 14% of children with epilepsy continue to experience seizures despite having appropriate medical therapy and develop medically refractory epilepsy (MRE) [8,9]. Assessing clinical outcomes and therapeutic efficacy in children with MRE undergoing palliative epilepsy surgery has been challenging because of the lack of a quantitative instrument capable of estimating the clinical status of these patients. The ideal instrument would at once consider seizure control, neurodevelopment, caregiver burden, and quality of life. The purpose of this study was to develop and pilot the Pediatric Refractory Epilepsy Questionnaire (PREQ), a quantitative instrument to assess the severity and individual burden of epilepsy in children with MRE undergoing palliative epilepsy treatments. Methods: The caregivers of 25 patients with MRE completed the PREQ and the Quality of Life in Childhood Epilepsy (QOLCE) measure and participated in a semistructured interview. Medical records of the patients were reviewed, an Early Childhood Epilepsy Severity Scale (E-CHESS) score was calculated, and a Global Assessment of Severity of Epilepsy (GASE) score was obtained for each patient. Key findings: The initial PREQ was modified based on the analysis of responses, association with previously validated scales, comments from caregivers, and expertise of the PREQ panelists. Pediatric Refractory Epilepsy Questionnaire subscale scores were calculated based on clinical paradigm and compared with independent measures of seizure severity and quality of life. Significant correlations were observed between the seizure severity subscale and the GASE score (r = 0.55) and between the mood subscale and the well-being score (r = 0.61) on the QOLCE. Significant correlations were also observed between the caregiver rating of seizure severity and the GASE score (r = 0.53), the social activity score (r = 0.57), and the behavior score (r = 0.43) on the QOLCE. Correlations between the caregiver rating of quality of life and the quality of life score (r = 0.58) and the number of AEDs used (r = 0.45) were also significant. Significance: This pilot study is an initial, critical step in the development of the PREQ. The significant correlations between the PREQ subscales and the external epilepsy severity and quality of life measures lend preliminary support to our hypothesis that the PREQ is assessing the severity of epilepsy along with other important domains, such as mood, neurodevelopment, and quality of life. A larger prospective study of this modified PREQ is currently underway to further develop the PREQ. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Epilepsy is a common neurological disorder in children [1], resulting in considerable morbidity, including cognitive, behavioral, and developmental impairment [2,3]. Furthermore, pediatric epilepsy is associated with increased mortality [4,5], substantial medicoeconomic burden ⁎ Corresponding author at: Division of Biology and Biomedical Sciences, Washington University in St. Louis Graduate School of Arts and Sciences, One Children's Place, Suite 4s20, St. Louis, MO 63110, USA. Tel.: +1 407 864 2546; fax: +1 314 454 2818. E-mail address: [email protected] (V. Purusothaman).
http://dx.doi.org/10.1016/j.yebeh.2014.04.025 1525-5050/© 2014 Elsevier Inc. All rights reserved.
[6,7], and incalculable human suffering. While pharmacotherapy is effective in many children with epilepsy, 6–14% of these children will continue to experience seizures despite having medical treatment [8,9]. Termed medically refractory epilepsy (MRE), this condition is characterized by failure to achieve seizure control after therapy with at least two appropriately chosen and appropriately dosed first-line antiepileptic drugs [10]. Children with MRE frequently have diffuse, poorly localized, or multifocal seizure onset, and treatment goals are often recalibrated to seizure palliation [11]. Assessing clinical outcomes and therapeutic efficacy in children with MRE is particularly challenging, as most traditional measures of
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
outcomes in patients with epilepsy focus on seizure frequency and severity [12–20]. In palliative epilepsy treatments, important considerations in assessing clinical status also include mood, behavior, schooling, neurodevelopment, medication side effects, and quality of life [21–26]. Several scales have been proposed to address these aspects but possess similar shortcomings as the traditional seizure severity scales [18,21,26–29]. Thus, there is a pressing need for a quantitative instrument capable of estimating the clinical status of patients with MRE undergoing palliative epilepsy treatments that at once considers seizure control, neurodevelopment, quality of life, and other issues. The purpose of this study was to develop and pilot the Pediatric Refractory Epilepsy Questionnaire (PREQ), a 22-question instrument completed by the child's caregiver. It is intended to assess the severity of epilepsy in children with MRE under consideration for palliative epilepsy treatments. The questionnaire focused on seizure severity, mood, behavior, development, schooling, and quality of life. The 22 items were initially developed using a literature review of current scales and input from clinicians and researchers relevant in the field of epilepsy and questionnaire development. The scale was then piloted with caregivers of 25 patients with MRE. Here, we present the results of the pilot phase of this study. 2. Methods 2.1. Questionnaire development Because of age- or disease-specific cognitive limitations in patients with MRE, the questionnaire was designed to be completed by caregivers of children with MRE. Health professionals with experience in dealing with MRE, including pediatric neurosurgeons, pediatric epileptologists, and pediatric neuropsychologists, were involved in identifying important parameters affecting the overall health status of a child with epilepsy. The following domains were identified as being important in assessing the status of a child with epilepsy: seizure severity, mood, behavior, development, medication requirements and side effects, impact on daily activities, schooling, caregiver burden, and quality of life. A list of items for the questionnaire was developed using two sources. Primarily, a review of the literature was performed to generate an inventory of items used in previous epilepsy severity and quality of life instruments. A team including clinicians, clinical researchers, and a questionnaire development expert was assembled to construct the individual questions and response options. 2.2. Pilot study 2.2.1. Study design and research subjects Following approval by the Human Research Protection Office of Washington University/St. Louis Children's Hospital (SLCH) (IRB # 201303059), data were collected from 25 subjects deemed to have MRE by an attending epileptologist. Patients with MRE and their families were approached in the outpatient neurology and neurosurgery clinics and on the video-EEG monitoring floor at SLCH. Participating caregivers began by completing the PREQ, and the time it required for completion was recorded. When more than one caregiver was present, both caregivers independently completed the PREQ. After completing the PREQ, caregivers participated in a semistructured interview to give feedback on the questionnaire. Caregivers were asked to comment on the clarity of the current questionnaire and identify question stems and answer choices that seemed confusing. Moreover, caregivers were also asked to determine whether the PREQ adequately addressed all aspects of their child's health and to identify additional aspects of their child's health that they thought should be included in the measure. A summary of their responses was recorded in writing. Following the interview, the caregivers filled out the Quality of Life in Childhood Epilepsy (QOLCE) measure [29]. Each subject's epileptologist then rated the severity of
27
the subject's epilepsy on the Global Assessment of Severity of Epilepsy (GASE) scale [21]. Finally, the medical records of each subject were reviewed, and data on demographics/clinical profile, seizure characteristics, use of medical care, pharmacotherapy, surgical treatment, and results of diagnostic studies (EEG and radiographic imaging) were collected. Review of medical records was limited to six months before patient enrollment, except to obtain data on surgical procedures and diagnostic studies. These data were then used to assign an Early Childhood Epilepsy Seizure Severity (E-CHESS) score for each subject [20]. 2.2.2. Statistical analysis Descriptive statistics were used to characterize demographic variables and interview responses. Data are presented as mean ± standard deviation for continuous variables and number of patients (% of group) for categorical variables. Descriptive statistics were also used to evaluate the responses to each item on the PREQ. To assess accuracy of caregiver recall, responses to select questions on the PREQ (seizure frequency and use of care) were correlated with corresponding data points obtained from medical records. Interrater reliability analysis was conducted using Pearson correlations (Fisher transformation). Interrater correlations were evaluated by comparing the lower bound of the 95% confidence interval to one. Questionnaire items were combined based on clinical context and a correlation matrix to create subscale scores for seizure severity, mood, impact on daily life, development, caregiver rating of seizure severity, and caregiver rating of quality of life. Cronbach's alpha was calculated to assess internal consistency and reliability of items on each subscale. These subscales were then evaluated for their association with age at onset of seizures, number of antiepileptic drugs (AEDs), MRI and EEG findings, GASE scores, E-CHESS scores, and QOLCE subscores. Pearson correlations (2-tailed t-test) were used to assess associations between subscales and continuous variables. Between-group comparisons of subscales were performed using analysis of variance (ANOVA). A p-value ≤ 0.05 was considered statistically significant. The data analysis was generated using SAS software (version 9.3) and SPSS software (version 19). 3. Results 3.1. Patient and seizure characteristics and interview responses Patient demographics and seizure characteristics are summarized in Table 1. Interview responses are summarized in Table 2. Table 1 Patient demographic and seizure characteristics. Total number of patients Female gender Age at seizure onset (in months) Age at time of survey (in months) Delay between seizure onset and time of survey (in months) Seizure type(s) GTC Tonic Clonic Atypical absence Atonic Myoclonic Simple partial seizures Complex partial seizures Infantile spasms Seizure frequency More than 10×/day Daily Weekly Monthly A few times in the past 6 months Epilepsy surgery VNS CC Partial Complete — 1 stage Complete — 2 stage
25 13 (52%) 19.6 ± 8.0 110.4 ± 13.5 88.6 ± 13.2 5 (20%) 11 (44%) 2 (8.0%) 4 (16%) 1 (4.0%) 3 (12%) 3 (12%) 9 (36%) 2 (8.0%) 5 (20%) 7 (28%) 8 (32%) 3 (12%) 1 (4.0%) 4 (16%) 3 (12%) 1 (4.0%) 1 (4.0%) 1 (4.0%)
Abbreviations: GTC, Generalized Tonic Clonic; VNS, Vagus Nerve Stimulation, CC, Corpus Callostomy.
28
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
Table 2 Interview responses. Confusions/concerns with current questionnaire Difficult to separate deficits due to epilepsy and other diagnoses Need a “never in the past 6 months” category Hard to remember 6 months/clarify 6 months in questions Need a “few times in the past 6 months” category Hard to distinguish between visits due to injury versus seizures Drop spells need clarifications Status — 1 continuous or clusters Needing a helmet depends on level of mobility Losing a job or changing work hours is not captured Questions do not apply for young children Other factors to be included in the questionnaire Life-threatening seizures Pain Medications and side effects Therapy types and effectiveness More questions on development
6 (24%) 2 (8.0%) 14 (56%) 6 (24%) 3 (12%) 4 (16%) 1 (4.0%) 3 (12%) 2 (8%) 4 (16%)
2 (8.0%) 1 (4.0%) 3 (12%) 1 (4.0%) 3 (12%)
3.2. PREQ responses A total of 31 questionnaires were completed. In the majority of cases (19/31, 76%), a single caregiver filled out the questionnaire, whereas two caregivers filled out the questionnaire for 6 (24%) patients. Among the respondents, 22 (71%) caregivers were mothers, seven (23%) were fathers, and two (6.5%) were grandmothers. It took the caregivers an average of 7.5 ± 3.0 min (range = 3.8–14.7 min) to complete the questionnaire. Response frequencies were calculated for each question. Most notably, on question 3a (frequency of urgent care visits for major injuries), 26 (84%) caregivers chose “never”, and five (16%) caregivers chose
“once a month”. No one chose the other four options. On the question asking about a seizure lasting longer that 15 min, only five (16%) caregivers out of 31 responded “yes”. On question 12 inquiring about mood and well-being, the number of “not applicable” responses made up a significant proportion of responses. Out of a total of 310 responses on question 12, 37 (12%) responses were N/A. While all 31 caregivers answered the Caregiver 1 question, there were 14 (45%) and 22 (71%) missing responses for the Caregiver 2 and 3 questions, respectively. Seizure frequency, number of seizure types, and number of seizurerelated emergency room (ER) and hospital admissions reported in the questionnaire were compared with corresponding values noted in the medical record. These relationships are presented in Fig. 1. Caregivers reported seizures as occurring less frequently on the PREQ than what was recorded in the medical record (Fig. 1A), whereas caregivers reported more seizure types on the PREQ than what was recorded in the medical record (Fig. 1B). In regard to seizure-related ER visits and admissions, there was high concordance between caregiver report on the PREQ and number of events noted in the medical record. When there was a discrepancy, caregivers reported more events than were noted in medical records (Figs. 1B–D). High interreliability was only seen for the seizure severity subscale (r = 0.98; 95% CI: 0.78 ≤ r ≤ 1.0). Because of the small number of patients with more than one respondent (n = 6), however, we were unable to demonstrate significant interrater reliability for the other subscores.
3.3. PREQ subscales and other disease severity measures The coefficients for the correlations between PREQ subscales and other disease severity measures are presented in Table 3. Significant
Fig. 1. PREQ responses compared with data from medical records. A. Seizure frequency on the PREQ versus the medical record — PREQ report of frequency is lower; B. Number of seizure types on the PREQ versus the medical record — PREQ report of seizure types is higher; C. Number of ER visits for seizures on the PREQ versus the medical record — PREQ report of ER visits is higher; D. Number of admissions for seizures on the PREQ versus the medical record — PREQ report of admissions is higher.
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
29
Table 3 PREQ subscale scores and other disease severity measures. Variable
Seizure severity
Mood
Impact on daily life
Development
Caregiver rating of seizure severity
Caregiver rating of quality of life
E-CHESS score GASE score QOLCE — QOL total QOLCE — physical activities QOLCE — well-being QOLCE — cognition QOLCE — social activities QOLCE — behavior QOLCE —general health QOLCE — quality of life Number of AEDs Age at seizure onset
0.19 0.55⁎⁎ −0.03 −0.07 −0.28 0.09 0.20 −0.16 0.32 0.37 0.24 −0.07
0.06 −0.09 −0.02 −0.32 0.61⁎⁎ 0.16 −0.17 −0.13 −0.22 0.02 −0.15 −0.04
−0.16 −0.15 0.06 −0.18 −0.11 −0.08 0.20 −0.17 0.34 0.33 0.00 −0.05
−0.02 0.04 0.44⁎ 0.49⁎ 0.33 0.25 0.11 0.29 0.27 0.28 0.20 −0.14
−0.02 0.53⁎ 0.46⁎ 0.30 0.08 0.04 0.57⁎⁎ 0.43⁎
0.11 0.04 −0.28 0.14 0.22 0.08 0.18 0.05 0.03 0.58⁎⁎ 0.45⁎ 0.15
0.20 0.37 0.21 −0.35
⁎ Correlation is significant at the 0.05 level (2-tailed). ⁎⁎ Correlation is significant at the 0.01 level (2-tailed).
correlations were observed between the seizure severity subscale and the GASE score (r = 0.55) and between the mood subscale and the well-being score (r = 0.61) on the QOLCE. Significant correlations were also observed between the caregiver rating of seizure severity and the GASE score (r = 0.53), the social activity score (r = 0.57), and the behavior score (r = 0.43) on the QOLCE. Correlations between the caregiver rating of quality of life and the quality of life score (r = 0.58) and the number of AEDs used (r = 0.45) were also significant. Cronbach's alpha for items on each of the seizure severity, mood, impact on daily life, and development subscales was 0.73, 0.81, 0.60, and 0.83, respectively. An exploratory analysis was performed to assess whether PREQ subscales differed between patients with different EEG findings and MRI findings. Because of the small number of patients, however, we were unable to observe a significant difference in the PREQ subscale scores between the different groups studied. Seven children who had undergone palliative epilepsy surgery (vagus nerve stimulation (VNS) or corpus callosotomy (CC)) were included in the analysis, but the number of patients was too small to make any conclusions regarding changes in PREQ subscales following surgery and the difference in PREQ scores between children who underwent surgery and those who did not. 3.4. Modification of the PREQ The response frequencies for each of the questions, along with the comments elicited during the interviews, were used to evaluate the questions and their associated scales. These data were presented to the PREQ panel of clinicians, clinical researchers, a neurodevelopmental psychologist, a statistician, and a questionnaire development expert, and final changes to the PREQ were made based on the expert opinion of this panel. Because all 31 respondents only used the first two options on the scale (“never” and “once a month”) on the question about major injuries sustained due to seizures, it was decided to expand the scale for that question. Only five (20%) patients had seizures that lasted longer than 15 min. Moreover, caregivers inquired about clusters of seizures when answering this question. Thus, it was decided to query seizures lasting longer than 5 min and clusters of seizures lasting longer than 15 min, as these parameters represent critical clinical time points that direct clinical care (timing of administration of rescue medication). It was also decided to remove the “N/A” option from question 12 in an attempt to decrease the number of missing responses. Instead, caregivers would be instructed both verbally and in the questionnaire to leave a question blank if their child was too young to experience the emotion. Finally, because of the large proportion of missing responses concerning the effects on the work life of additional caregivers (Caregiver 2 and Caregiver 3), it was decided to only ask about the caregiver (Caregiver 1) completing the questionnaire.
The questionnaire was also modified based on the interview comments (Table 2). To help clarify the time frame of six months, the instructions were made more noticeable, and the time frame was repeated throughout the questionnaire. Drop seizures were defined. The question about protective devices was expanded to query whether patients used other assistive devices, such as wheelchairs and special needs strollers. An additional question was added to address changes in caregiver employment, such as changing work hours, choosing to work from home or stay at home, and losing a job. The question about schooling was changed to include options that would be applicable for very young children [18]. This modified copy of the PREQ can be found in the supplemental section.
4. Discussion This pilot study represents an initial, critical step in the development of the PREQ. Responses for each question were analyzed statistically and then reviewed with a panel of pediatric epileptologists and neurosurgeons, a neurodevelopmental psychologist, a statistician, and an expert in questionnaire development. The initial PREQ was modified based on the analysis of responses, association with previously validated scales, comments from questionnaire interviewees, and expertise of the PREQ panelists. There are several key challenges in developing a relevant questionnaire for assessing the individual burden of MRE. One concern identified early in the development of the PREQ was difficulty in assessing whether the seizure disorder or another comorbid disorder is the cause for any given deficit or problem. Thus, the revised PREQ stresses to the caregiver the importance of considering the child as a whole. Another important issue surrounds AEDs, their dosing, and their adverse effects. Cleary, large doses of AEDs can affect mood, behavior, school performance, development, and other issues; however, a validated and clinically useful questionnaire looking at medication side effects, the Pediatric Epilepsy Side Effects Questionnaire (PESQ) [30,31], already exists. To reduce redundancy, we anticipate that the PREQ will be used in tandem with the PESQ. While there are several established scales to assess a child's developmental age, these questionnaires are long and often need to be tailored for each individual child by a trained clinician [32,33]. The PREQ was designed as a caregiver rating system to asses the child's development, thus, reducing the time and personnel required to assess this domain. In the larger study of the PREQ, the Adaptive Behavior Assessment System — Second Edition (ABAS-II) and the Behavior Assessment System for Children — Second Edition (BASC-II) are administered to confirm that the caregiver rating system is reliable and valid in the setting of the PREQ.
30
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
When comparing the questionnaire responses with data obtained from the medical record, it was noted that the caregivers reported seizures as occurring less often than were noted by the neurologists. One reason for this discrepancy could be that the caregivers are averaging over the past six months as asked by the PREQ, while the physicians are recording the current frequency. The caregivers also reported more total types of seizures. This discrepancy is most likely due to differing definition for types of seizures. During the review of medical records, the number of seizure types was calculated based on identified seizure types such as generalized tonic–clonic and complex partial seizures. If the caregivers used their interpretation of the different seizure semiologies they see in their child, it is possible that they overestimated the types of seizures. The majority of the caregiver responses on the number of ER visits and admissions were concordant with the medical record. When there were discrepancies, caregivers reported more events on the PREQ, suggesting that we were able to capture more information with the PREQ than we otherwise would with just a retrospective review of the medical record. The correlations between the subscales calculated from the PREQ and the external epilepsy severity and quality of life measures lend preliminary support to our hypothesis that our questionnaire is measuring the severity of epilepsy along with other domains identified to be important such as impact on daily activities, mood, schooling, development, and quality of life. There is, however, an inherent limitation to using a clinical paradigm to create subscores. Questionnaire development often employs principal component analysis (PCA) to assess the inherent structure of the data and determine if questions associate in a manner that matches the clinical paradigm [20,21,34]. However, because of the small sample size of this pilot study (n = 25), the clustering of items based on face validity that we report here requires replication using principal components analysis (PCA) in a larger sample to further evaluate the underlying factor structure of items we developed for this survey and the internal consistency of items that load on specific factors [35]. Moreover, internal validity and construct validity should be assessed in a larger study. Internal validity will be examined via test–retest and interrater reliability estimation [20,21,36]. Construct validity for PREQ will be investigated by correlating the scores on the PREQ with GASE, E-CHESS, QOLCE, ABAS-II, and BASC-II. Clinical indicators abstracted from the medical records will also be used to assess construct validity. Factors that have been shown to predict MRE in childhood epilepsy include age at onset, multifocal spikes and generalized slowing on EEG, abnormal MRI, multiple seizure types, and mental and motor deficits [8,10,37–40]. In validating this questionnaire, it will be important to address the pitfalls faced by the current instruments. Many of the currently available instruments to assess seizure severity such as the Hague Seizure Severity Scale for Children and the E-CHESS are limited [14,15,20,41]. The E-CHESS was developed and validated for use in infants [20]. The Hague, while catering to a broader age range, is fraught with challenges regarding clinical relevance, scaling issues, and responsiveness of the scale in relation to clinical changes [14,15,41]. These difficulties with clinical relevance, scaling issues, and responsiveness of scale in relation to clinical changes are also seen with the QOLCE [12,26,28,29]. Thus, it will be important to assess the responsiveness of the PREQ to change and establish clinical relevance of the scores computed from the PREQ as future studies are carried out to develop the questionnaire further. 5. Conclusion This pilot study is an initial, critical step in the development of the PREQ. The significant correlations between the PREQ subscale scores and the external epilepsy severity and quality of life measures lend preliminary support to our hypothesis that the PREQ is assessing the severity of epilepsy along with other important domains, such as mood, development, and quality of life. A larger prospective study of
this modified PREQ is currently underway to further develop the PREQ and validate it as a quantitative instrument that can be used to reliably assess the clinical status of children with MRE undergoing palliative epilepsy treatment. Acknowledgments The authors thank Larry Snyder, M.D., Ph.D. and Jay F. Piccirillo, M.D. for their help in research design. The authors thank Janice E. BrunstromHernandez, M.D., Michael A. Ciliberto, M.D., Michael J. Noetzel, M.D., John M. Zempel, M.D., Ph.D., Liu-Lin Thio, M.D., Ph.D., Judy Weisenberg, M.D., Michael Wong, M.D., Ph.D., Rohini Coorg, M.D., Bryan McGill, M.D., Ph.D., and Adam Ostendorf, M.D. for their help in patient recruitment and participation in the study. The authors thank Deanna Mercer for her help with Institutional Board Review. This research was funded by the Washington University School of Medicine Fund. Conflict of interest statement None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2014.04.025. References [1] Shinnar S, Pellock JM. Update on the epidemiology and prognosis of pediatric epilepsy. J Child Neurol 2002;17(Suppl. 1):S4–S17. [2] Bailet LL, Turk WR. The impact of childhood epilepsy on neurocognitive and behavioral performance: a prospective longitudinal study. Epilepsia 2000;41(4):426–31. [3] Bjornaes H, et al. The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study. Seizure 2001;10(4):250–9. [4] Callenbach PM, Westendorp RG, Geerts AT, Arts WF, Peeters EA, van Donselaar CA, et al. Mortality risk in children with epilepsy: the Dutch study of epilepsy in childhood. Pediatrics 2001;107(6):1259–63. [5] Breningstall GN. Mortality in pediatric epilepsy. Pediatr Neurol 2001;25(1):9–16. [6] Strzelczyk A, Reese JP, Dodel R, Hamer HM, Callenbach PM. Cost of epilepsy: a systematic review. Pharmacoeconomics 2008;26(6):463–76. [7] O'Dell C, Wheless JW, Cloyd J. The personal and financial impact of repetitive or prolonged seizures on the patient and family. J Child Neurol 2007;22(5 Suppl.): 61S–70S. [8] Seker Yilmaz B, Okuyaz C, Komur M. Predictors of intractable childhood epilepsy. Pediatr Neurol 2013;48(1):52–5. [9] Pressler RM. Vagus nerve stimulation in children with intractable epilepsy. Dev Med Child Neurol 2012;54(9):782–3. [10] Berg AT, Shinnar S, Levy SR, Testa FM, Smith-Rapaport S, Beckerman B. Early development of intractable epilepsy in children: a prospective study. Neurology 2001;56(11):1445–52. [11] Nei M, O'Connor M, Liporace J, Sperling MR, Callenbach PM, Callenbach PM. Refractory generalized seizures: response to corpus callosotomy and vagal nerve stimulation. Epilepsia 2006;47(1):115–22. [12] Sabaz M, Lawson JA, Cairns DR, Duchowny MS, Resnick TJ, Dean PM, et al. The impact of epilepsy surgery on quality of life in children. Neurology 2006;66(4):557–61. [13] Baker GA, Smith DF, Jacoby A, Hayes JA, Chadwick DW. Liverpool Seizure Severity Scale revisited. Seizure 1998;7(3):201–5. [14] Cramer JA. Assessing the severity of seizures and epilepsy: which scales are valid? Curr Opin Neurol 2001;14(2):225–9. [15] Carpay JA, Vermuelen J, Stroink H, Brouwer OF, Peters AC, Aldenkamp AP, et al. Seizure severity in children with epilepsy: a parent-completed scale compared with clinical data. Epilepsia 1997;38(3):346–52. [16] Engel Jr J, E. International League Against. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001;42(6):796–803. [17] Wieser HG, Blume WT, Fish D, Goldensohn E, Hufnagel A, King D, et al. ILAE Commission Report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001;42(2):282–6. [18] Camfield C, Breau L, Camfield P. Impact of pediatric epilepsy on the family: a new scale for clinical and research use. Epilepsia 2001;42(1):104–12. [19] O'Donoghue MF, Duncan JS, Sander JW. The National Hospital Seizure Severity Scale: a further development of the Chalfont Seizure Severity Scale. Epilepsia 1996;37(6): 563–71.
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31 [20] Humphrey A, Ploubidis GB, Yates JR, Steinberg T, Bolton PF, Callenbach PM. The Early Childhood Epilepsy Severity Scale (E-Chess). Epilepsy Res 2008;79(2–3):139–45. [21] Speechley KN, Sang X, Levin S, Zou GY, Eliasziw M, Smith M, et al. Assessing severity of epilepsy in children: preliminary evidence of validity and reliability of a singleitem scale. Epilepsy Behav 2008;13(2):337–42. [22] Kanner AM. The complex epilepsy patient: intricacies of assessment and treatment. Epilepsia 2003;44(Suppl. 5):3–8. [23] Pellock JM. Defining the problem: psychiatric and behavioral comorbidity in children and adolescents with epilepsy. Epilepsy Behav 2004;5(Suppl. 3):S3–9. [24] Loring DW, Meador KJ, Lee GP. Determinants of quality of life in epilepsy. Epilepsy Behav 2004;5(6):976–80. [25] Wheless JW. Intractable epilepsy: a survey of patients and caregivers. Epilepsy Behav 2006;8(4):756–64. [26] Titus JB, Lee A, Kasasbeth A, Thio LL, Stephenson J, Steger-May K, et al. Health-related quality of life before and after pediatric epilepsy surgery: the influence of seizure outcome on changes in physical functioning and social functioning. Epilepsy Behav 2013;27(3):477–83. [27] Speechley KN, Ferro MA, Camfield CS, Huang W, Levin SD, Smith ML, et al. Quality of life in children with new-onset epilepsy: a 2-year prospective cohort study. Neurology 2012;79(15):1548–55. [28] Tanriverdi T, Olivier NP, Olivier A. Quality of life after extratemporal epilepsy surgery: a prospective clinical study. Clin Neurol Neurosurg 2008;110(1):30–7. [29] Sabaz M, Lawson JA, Cairns DR, Duchowny MS, Resnick TJ, Dean PM, et al. Validation of the Quality of Life in Childhood Epilepsy Questionnaire in American epilepsy patients. Epilepsy Behav 2003;4(6):680–91. [30] Morita DA, Glauser TA, Modi AC. Development and validation of the Pediatric Epilepsy Side Effects Questionnaire. Neurology 2012;79(12):1252–8.
31
[31] Kothare SV, Wagner J. Quality of life in pediatric epilepsy: a validated questionnaire for side effects of AEDs. Neurology 2012;79(12):1194–5. [32] Wei Y, Oakland T, Algina J. Multigroup confirmatory factor analysis for the adaptive behavior assessment system-II parent form, ages 5–21. Am J Ment Retard 2008;113(3):178–86. [33] Vohr BR, Stephens BE, Higgins RD, Bann CM, Hintz SR, Das A, et al. Are outcomes of extremely preterm infants improving? Impact of Bayley assessment on outcomes. J Pediatr 2012;161(2):222–8 [e3]. [34] Fayers PM, Groenvold M, Hand DJ, Bjordal K. Clinical impact versus factor analysis for quality of life questionnaire construction. J Clin Epidemiol 1998;51(3):285–6. [35] Lautenschlager GJ. A comparison of alternatives to conducting Monte-Carlo analyses for determining parallel analysis criteria. Multivar Behav Res 1989;24(3): 365–95. [36] Kulkarni AV, Rabin D, Drake JM. An instrument to measure the health status in children with hydrocephalus: the Hydrocephalus Outcome Questionnaire. J Neurosurg 2004;101(2 Suppl.):134–40. [37] Ko TS, Holmes GL. EEG and clinical predictors of medically intractable childhood epilepsy. Clin Neurophysiol 1999;110(7):1245–51. [38] Ohtsuka Y, Yoshinaga H, Kobayashi K. Refractory childhood epilepsy and factors related to refractoriness. Epilepsia 2000;41(Suppl. 9):14–7. [39] Chawla S, Aneja S, Kashyap R, Mallika V. Etiology and clinical predictors of intractable epilepsy. Pediatr Neurol 2002;27(3):186–91. [40] Berg AT, Levy SR, Novotny EJ, Shinnar S. Predictors of intractable epilepsy in childhood: a case–control study. Epilepsia 1996;37(1):24–30. [41] Cramer JA, French J. Quantitative assessment of seizure severity for clinical trials: a review of approaches to seizure components. Epilepsia 2001;42(1):119–29.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Developing the Pediatric Refractory Epilepsy Questionnaire: A pilot study Vaishnavi Purusothaman a,⁎, Robin C.C. Ryther b, Mary Bertrand b, Lisa A. Harker c, Donna B. Jeffe d, Michael Wallendorf e, Matthew D. Smyth f, David D. Limbrick f a
Division of Biology and Biomedical Sciences, Washington University in St. Louis Graduate School of Arts and Sciences, St. Louis, MO, USA Department of Neurology, Saint Louis Children's Hospital, Washington University in St. Louis School of Medicine, St. Louis, MO, USA Department of Psychology, Saint Louis Children's Hospital, Washington University in St. Louis School of Medicine in St. Louis, St. Louis, MO, USA d Division of General Medical Sciences, Department of Medicine, Washington University in St. Louis, School of Medicine St. Louis, MO, USA e Division of Biostatistics, Washington University in St. Louis School of Medicine in St. Louis, St. Louis, MO, USA f Department of Neurological Surgery, Saint Louis Children's Hospital, Washington University in St. Louis, School of Medicine St. Louis, MO, USA b c
a r t i c l e
i n f o
Article history: Received 5 January 2014 Revised 25 April 2014 Accepted 29 April 2014 Available online xxxx Keywords: Medically refractory Intractable Seizure Epilepsy Questionnaire Epilepsy surgery Palliative epilepsy surgery
a b s t r a c t Purpose: Up to 14% of children with epilepsy continue to experience seizures despite having appropriate medical therapy and develop medically refractory epilepsy (MRE) [8,9]. Assessing clinical outcomes and therapeutic efficacy in children with MRE undergoing palliative epilepsy surgery has been challenging because of the lack of a quantitative instrument capable of estimating the clinical status of these patients. The ideal instrument would at once consider seizure control, neurodevelopment, caregiver burden, and quality of life. The purpose of this study was to develop and pilot the Pediatric Refractory Epilepsy Questionnaire (PREQ), a quantitative instrument to assess the severity and individual burden of epilepsy in children with MRE undergoing palliative epilepsy treatments. Methods: The caregivers of 25 patients with MRE completed the PREQ and the Quality of Life in Childhood Epilepsy (QOLCE) measure and participated in a semistructured interview. Medical records of the patients were reviewed, an Early Childhood Epilepsy Severity Scale (E-CHESS) score was calculated, and a Global Assessment of Severity of Epilepsy (GASE) score was obtained for each patient. Key findings: The initial PREQ was modified based on the analysis of responses, association with previously validated scales, comments from caregivers, and expertise of the PREQ panelists. Pediatric Refractory Epilepsy Questionnaire subscale scores were calculated based on clinical paradigm and compared with independent measures of seizure severity and quality of life. Significant correlations were observed between the seizure severity subscale and the GASE score (r = 0.55) and between the mood subscale and the well-being score (r = 0.61) on the QOLCE. Significant correlations were also observed between the caregiver rating of seizure severity and the GASE score (r = 0.53), the social activity score (r = 0.57), and the behavior score (r = 0.43) on the QOLCE. Correlations between the caregiver rating of quality of life and the quality of life score (r = 0.58) and the number of AEDs used (r = 0.45) were also significant. Significance: This pilot study is an initial, critical step in the development of the PREQ. The significant correlations between the PREQ subscales and the external epilepsy severity and quality of life measures lend preliminary support to our hypothesis that the PREQ is assessing the severity of epilepsy along with other important domains, such as mood, neurodevelopment, and quality of life. A larger prospective study of this modified PREQ is currently underway to further develop the PREQ. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Epilepsy is a common neurological disorder in children [1], resulting in considerable morbidity, including cognitive, behavioral, and developmental impairment [2,3]. Furthermore, pediatric epilepsy is associated with increased mortality [4,5], substantial medicoeconomic burden ⁎ Corresponding author at: Division of Biology and Biomedical Sciences, Washington University in St. Louis Graduate School of Arts and Sciences, One Children's Place, Suite 4s20, St. Louis, MO 63110, USA. Tel.: +1 407 864 2546; fax: +1 314 454 2818. E-mail address: [email protected] (V. Purusothaman).
http://dx.doi.org/10.1016/j.yebeh.2014.04.025 1525-5050/© 2014 Elsevier Inc. All rights reserved.
[6,7], and incalculable human suffering. While pharmacotherapy is effective in many children with epilepsy, 6–14% of these children will continue to experience seizures despite having medical treatment [8,9]. Termed medically refractory epilepsy (MRE), this condition is characterized by failure to achieve seizure control after therapy with at least two appropriately chosen and appropriately dosed first-line antiepileptic drugs [10]. Children with MRE frequently have diffuse, poorly localized, or multifocal seizure onset, and treatment goals are often recalibrated to seizure palliation [11]. Assessing clinical outcomes and therapeutic efficacy in children with MRE is particularly challenging, as most traditional measures of
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
outcomes in patients with epilepsy focus on seizure frequency and severity [12–20]. In palliative epilepsy treatments, important considerations in assessing clinical status also include mood, behavior, schooling, neurodevelopment, medication side effects, and quality of life [21–26]. Several scales have been proposed to address these aspects but possess similar shortcomings as the traditional seizure severity scales [18,21,26–29]. Thus, there is a pressing need for a quantitative instrument capable of estimating the clinical status of patients with MRE undergoing palliative epilepsy treatments that at once considers seizure control, neurodevelopment, quality of life, and other issues. The purpose of this study was to develop and pilot the Pediatric Refractory Epilepsy Questionnaire (PREQ), a 22-question instrument completed by the child's caregiver. It is intended to assess the severity of epilepsy in children with MRE under consideration for palliative epilepsy treatments. The questionnaire focused on seizure severity, mood, behavior, development, schooling, and quality of life. The 22 items were initially developed using a literature review of current scales and input from clinicians and researchers relevant in the field of epilepsy and questionnaire development. The scale was then piloted with caregivers of 25 patients with MRE. Here, we present the results of the pilot phase of this study. 2. Methods 2.1. Questionnaire development Because of age- or disease-specific cognitive limitations in patients with MRE, the questionnaire was designed to be completed by caregivers of children with MRE. Health professionals with experience in dealing with MRE, including pediatric neurosurgeons, pediatric epileptologists, and pediatric neuropsychologists, were involved in identifying important parameters affecting the overall health status of a child with epilepsy. The following domains were identified as being important in assessing the status of a child with epilepsy: seizure severity, mood, behavior, development, medication requirements and side effects, impact on daily activities, schooling, caregiver burden, and quality of life. A list of items for the questionnaire was developed using two sources. Primarily, a review of the literature was performed to generate an inventory of items used in previous epilepsy severity and quality of life instruments. A team including clinicians, clinical researchers, and a questionnaire development expert was assembled to construct the individual questions and response options. 2.2. Pilot study 2.2.1. Study design and research subjects Following approval by the Human Research Protection Office of Washington University/St. Louis Children's Hospital (SLCH) (IRB # 201303059), data were collected from 25 subjects deemed to have MRE by an attending epileptologist. Patients with MRE and their families were approached in the outpatient neurology and neurosurgery clinics and on the video-EEG monitoring floor at SLCH. Participating caregivers began by completing the PREQ, and the time it required for completion was recorded. When more than one caregiver was present, both caregivers independently completed the PREQ. After completing the PREQ, caregivers participated in a semistructured interview to give feedback on the questionnaire. Caregivers were asked to comment on the clarity of the current questionnaire and identify question stems and answer choices that seemed confusing. Moreover, caregivers were also asked to determine whether the PREQ adequately addressed all aspects of their child's health and to identify additional aspects of their child's health that they thought should be included in the measure. A summary of their responses was recorded in writing. Following the interview, the caregivers filled out the Quality of Life in Childhood Epilepsy (QOLCE) measure [29]. Each subject's epileptologist then rated the severity of
27
the subject's epilepsy on the Global Assessment of Severity of Epilepsy (GASE) scale [21]. Finally, the medical records of each subject were reviewed, and data on demographics/clinical profile, seizure characteristics, use of medical care, pharmacotherapy, surgical treatment, and results of diagnostic studies (EEG and radiographic imaging) were collected. Review of medical records was limited to six months before patient enrollment, except to obtain data on surgical procedures and diagnostic studies. These data were then used to assign an Early Childhood Epilepsy Seizure Severity (E-CHESS) score for each subject [20]. 2.2.2. Statistical analysis Descriptive statistics were used to characterize demographic variables and interview responses. Data are presented as mean ± standard deviation for continuous variables and number of patients (% of group) for categorical variables. Descriptive statistics were also used to evaluate the responses to each item on the PREQ. To assess accuracy of caregiver recall, responses to select questions on the PREQ (seizure frequency and use of care) were correlated with corresponding data points obtained from medical records. Interrater reliability analysis was conducted using Pearson correlations (Fisher transformation). Interrater correlations were evaluated by comparing the lower bound of the 95% confidence interval to one. Questionnaire items were combined based on clinical context and a correlation matrix to create subscale scores for seizure severity, mood, impact on daily life, development, caregiver rating of seizure severity, and caregiver rating of quality of life. Cronbach's alpha was calculated to assess internal consistency and reliability of items on each subscale. These subscales were then evaluated for their association with age at onset of seizures, number of antiepileptic drugs (AEDs), MRI and EEG findings, GASE scores, E-CHESS scores, and QOLCE subscores. Pearson correlations (2-tailed t-test) were used to assess associations between subscales and continuous variables. Between-group comparisons of subscales were performed using analysis of variance (ANOVA). A p-value ≤ 0.05 was considered statistically significant. The data analysis was generated using SAS software (version 9.3) and SPSS software (version 19). 3. Results 3.1. Patient and seizure characteristics and interview responses Patient demographics and seizure characteristics are summarized in Table 1. Interview responses are summarized in Table 2. Table 1 Patient demographic and seizure characteristics. Total number of patients Female gender Age at seizure onset (in months) Age at time of survey (in months) Delay between seizure onset and time of survey (in months) Seizure type(s) GTC Tonic Clonic Atypical absence Atonic Myoclonic Simple partial seizures Complex partial seizures Infantile spasms Seizure frequency More than 10×/day Daily Weekly Monthly A few times in the past 6 months Epilepsy surgery VNS CC Partial Complete — 1 stage Complete — 2 stage
25 13 (52%) 19.6 ± 8.0 110.4 ± 13.5 88.6 ± 13.2 5 (20%) 11 (44%) 2 (8.0%) 4 (16%) 1 (4.0%) 3 (12%) 3 (12%) 9 (36%) 2 (8.0%) 5 (20%) 7 (28%) 8 (32%) 3 (12%) 1 (4.0%) 4 (16%) 3 (12%) 1 (4.0%) 1 (4.0%) 1 (4.0%)
Abbreviations: GTC, Generalized Tonic Clonic; VNS, Vagus Nerve Stimulation, CC, Corpus Callostomy.
28
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
Table 2 Interview responses. Confusions/concerns with current questionnaire Difficult to separate deficits due to epilepsy and other diagnoses Need a “never in the past 6 months” category Hard to remember 6 months/clarify 6 months in questions Need a “few times in the past 6 months” category Hard to distinguish between visits due to injury versus seizures Drop spells need clarifications Status — 1 continuous or clusters Needing a helmet depends on level of mobility Losing a job or changing work hours is not captured Questions do not apply for young children Other factors to be included in the questionnaire Life-threatening seizures Pain Medications and side effects Therapy types and effectiveness More questions on development
6 (24%) 2 (8.0%) 14 (56%) 6 (24%) 3 (12%) 4 (16%) 1 (4.0%) 3 (12%) 2 (8%) 4 (16%)
2 (8.0%) 1 (4.0%) 3 (12%) 1 (4.0%) 3 (12%)
3.2. PREQ responses A total of 31 questionnaires were completed. In the majority of cases (19/31, 76%), a single caregiver filled out the questionnaire, whereas two caregivers filled out the questionnaire for 6 (24%) patients. Among the respondents, 22 (71%) caregivers were mothers, seven (23%) were fathers, and two (6.5%) were grandmothers. It took the caregivers an average of 7.5 ± 3.0 min (range = 3.8–14.7 min) to complete the questionnaire. Response frequencies were calculated for each question. Most notably, on question 3a (frequency of urgent care visits for major injuries), 26 (84%) caregivers chose “never”, and five (16%) caregivers chose
“once a month”. No one chose the other four options. On the question asking about a seizure lasting longer that 15 min, only five (16%) caregivers out of 31 responded “yes”. On question 12 inquiring about mood and well-being, the number of “not applicable” responses made up a significant proportion of responses. Out of a total of 310 responses on question 12, 37 (12%) responses were N/A. While all 31 caregivers answered the Caregiver 1 question, there were 14 (45%) and 22 (71%) missing responses for the Caregiver 2 and 3 questions, respectively. Seizure frequency, number of seizure types, and number of seizurerelated emergency room (ER) and hospital admissions reported in the questionnaire were compared with corresponding values noted in the medical record. These relationships are presented in Fig. 1. Caregivers reported seizures as occurring less frequently on the PREQ than what was recorded in the medical record (Fig. 1A), whereas caregivers reported more seizure types on the PREQ than what was recorded in the medical record (Fig. 1B). In regard to seizure-related ER visits and admissions, there was high concordance between caregiver report on the PREQ and number of events noted in the medical record. When there was a discrepancy, caregivers reported more events than were noted in medical records (Figs. 1B–D). High interreliability was only seen for the seizure severity subscale (r = 0.98; 95% CI: 0.78 ≤ r ≤ 1.0). Because of the small number of patients with more than one respondent (n = 6), however, we were unable to demonstrate significant interrater reliability for the other subscores.
3.3. PREQ subscales and other disease severity measures The coefficients for the correlations between PREQ subscales and other disease severity measures are presented in Table 3. Significant
Fig. 1. PREQ responses compared with data from medical records. A. Seizure frequency on the PREQ versus the medical record — PREQ report of frequency is lower; B. Number of seizure types on the PREQ versus the medical record — PREQ report of seizure types is higher; C. Number of ER visits for seizures on the PREQ versus the medical record — PREQ report of ER visits is higher; D. Number of admissions for seizures on the PREQ versus the medical record — PREQ report of admissions is higher.
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
29
Table 3 PREQ subscale scores and other disease severity measures. Variable
Seizure severity
Mood
Impact on daily life
Development
Caregiver rating of seizure severity
Caregiver rating of quality of life
E-CHESS score GASE score QOLCE — QOL total QOLCE — physical activities QOLCE — well-being QOLCE — cognition QOLCE — social activities QOLCE — behavior QOLCE —general health QOLCE — quality of life Number of AEDs Age at seizure onset
0.19 0.55⁎⁎ −0.03 −0.07 −0.28 0.09 0.20 −0.16 0.32 0.37 0.24 −0.07
0.06 −0.09 −0.02 −0.32 0.61⁎⁎ 0.16 −0.17 −0.13 −0.22 0.02 −0.15 −0.04
−0.16 −0.15 0.06 −0.18 −0.11 −0.08 0.20 −0.17 0.34 0.33 0.00 −0.05
−0.02 0.04 0.44⁎ 0.49⁎ 0.33 0.25 0.11 0.29 0.27 0.28 0.20 −0.14
−0.02 0.53⁎ 0.46⁎ 0.30 0.08 0.04 0.57⁎⁎ 0.43⁎
0.11 0.04 −0.28 0.14 0.22 0.08 0.18 0.05 0.03 0.58⁎⁎ 0.45⁎ 0.15
0.20 0.37 0.21 −0.35
⁎ Correlation is significant at the 0.05 level (2-tailed). ⁎⁎ Correlation is significant at the 0.01 level (2-tailed).
correlations were observed between the seizure severity subscale and the GASE score (r = 0.55) and between the mood subscale and the well-being score (r = 0.61) on the QOLCE. Significant correlations were also observed between the caregiver rating of seizure severity and the GASE score (r = 0.53), the social activity score (r = 0.57), and the behavior score (r = 0.43) on the QOLCE. Correlations between the caregiver rating of quality of life and the quality of life score (r = 0.58) and the number of AEDs used (r = 0.45) were also significant. Cronbach's alpha for items on each of the seizure severity, mood, impact on daily life, and development subscales was 0.73, 0.81, 0.60, and 0.83, respectively. An exploratory analysis was performed to assess whether PREQ subscales differed between patients with different EEG findings and MRI findings. Because of the small number of patients, however, we were unable to observe a significant difference in the PREQ subscale scores between the different groups studied. Seven children who had undergone palliative epilepsy surgery (vagus nerve stimulation (VNS) or corpus callosotomy (CC)) were included in the analysis, but the number of patients was too small to make any conclusions regarding changes in PREQ subscales following surgery and the difference in PREQ scores between children who underwent surgery and those who did not. 3.4. Modification of the PREQ The response frequencies for each of the questions, along with the comments elicited during the interviews, were used to evaluate the questions and their associated scales. These data were presented to the PREQ panel of clinicians, clinical researchers, a neurodevelopmental psychologist, a statistician, and a questionnaire development expert, and final changes to the PREQ were made based on the expert opinion of this panel. Because all 31 respondents only used the first two options on the scale (“never” and “once a month”) on the question about major injuries sustained due to seizures, it was decided to expand the scale for that question. Only five (20%) patients had seizures that lasted longer than 15 min. Moreover, caregivers inquired about clusters of seizures when answering this question. Thus, it was decided to query seizures lasting longer than 5 min and clusters of seizures lasting longer than 15 min, as these parameters represent critical clinical time points that direct clinical care (timing of administration of rescue medication). It was also decided to remove the “N/A” option from question 12 in an attempt to decrease the number of missing responses. Instead, caregivers would be instructed both verbally and in the questionnaire to leave a question blank if their child was too young to experience the emotion. Finally, because of the large proportion of missing responses concerning the effects on the work life of additional caregivers (Caregiver 2 and Caregiver 3), it was decided to only ask about the caregiver (Caregiver 1) completing the questionnaire.
The questionnaire was also modified based on the interview comments (Table 2). To help clarify the time frame of six months, the instructions were made more noticeable, and the time frame was repeated throughout the questionnaire. Drop seizures were defined. The question about protective devices was expanded to query whether patients used other assistive devices, such as wheelchairs and special needs strollers. An additional question was added to address changes in caregiver employment, such as changing work hours, choosing to work from home or stay at home, and losing a job. The question about schooling was changed to include options that would be applicable for very young children [18]. This modified copy of the PREQ can be found in the supplemental section.
4. Discussion This pilot study represents an initial, critical step in the development of the PREQ. Responses for each question were analyzed statistically and then reviewed with a panel of pediatric epileptologists and neurosurgeons, a neurodevelopmental psychologist, a statistician, and an expert in questionnaire development. The initial PREQ was modified based on the analysis of responses, association with previously validated scales, comments from questionnaire interviewees, and expertise of the PREQ panelists. There are several key challenges in developing a relevant questionnaire for assessing the individual burden of MRE. One concern identified early in the development of the PREQ was difficulty in assessing whether the seizure disorder or another comorbid disorder is the cause for any given deficit or problem. Thus, the revised PREQ stresses to the caregiver the importance of considering the child as a whole. Another important issue surrounds AEDs, their dosing, and their adverse effects. Cleary, large doses of AEDs can affect mood, behavior, school performance, development, and other issues; however, a validated and clinically useful questionnaire looking at medication side effects, the Pediatric Epilepsy Side Effects Questionnaire (PESQ) [30,31], already exists. To reduce redundancy, we anticipate that the PREQ will be used in tandem with the PESQ. While there are several established scales to assess a child's developmental age, these questionnaires are long and often need to be tailored for each individual child by a trained clinician [32,33]. The PREQ was designed as a caregiver rating system to asses the child's development, thus, reducing the time and personnel required to assess this domain. In the larger study of the PREQ, the Adaptive Behavior Assessment System — Second Edition (ABAS-II) and the Behavior Assessment System for Children — Second Edition (BASC-II) are administered to confirm that the caregiver rating system is reliable and valid in the setting of the PREQ.
30
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31
When comparing the questionnaire responses with data obtained from the medical record, it was noted that the caregivers reported seizures as occurring less often than were noted by the neurologists. One reason for this discrepancy could be that the caregivers are averaging over the past six months as asked by the PREQ, while the physicians are recording the current frequency. The caregivers also reported more total types of seizures. This discrepancy is most likely due to differing definition for types of seizures. During the review of medical records, the number of seizure types was calculated based on identified seizure types such as generalized tonic–clonic and complex partial seizures. If the caregivers used their interpretation of the different seizure semiologies they see in their child, it is possible that they overestimated the types of seizures. The majority of the caregiver responses on the number of ER visits and admissions were concordant with the medical record. When there were discrepancies, caregivers reported more events on the PREQ, suggesting that we were able to capture more information with the PREQ than we otherwise would with just a retrospective review of the medical record. The correlations between the subscales calculated from the PREQ and the external epilepsy severity and quality of life measures lend preliminary support to our hypothesis that our questionnaire is measuring the severity of epilepsy along with other domains identified to be important such as impact on daily activities, mood, schooling, development, and quality of life. There is, however, an inherent limitation to using a clinical paradigm to create subscores. Questionnaire development often employs principal component analysis (PCA) to assess the inherent structure of the data and determine if questions associate in a manner that matches the clinical paradigm [20,21,34]. However, because of the small sample size of this pilot study (n = 25), the clustering of items based on face validity that we report here requires replication using principal components analysis (PCA) in a larger sample to further evaluate the underlying factor structure of items we developed for this survey and the internal consistency of items that load on specific factors [35]. Moreover, internal validity and construct validity should be assessed in a larger study. Internal validity will be examined via test–retest and interrater reliability estimation [20,21,36]. Construct validity for PREQ will be investigated by correlating the scores on the PREQ with GASE, E-CHESS, QOLCE, ABAS-II, and BASC-II. Clinical indicators abstracted from the medical records will also be used to assess construct validity. Factors that have been shown to predict MRE in childhood epilepsy include age at onset, multifocal spikes and generalized slowing on EEG, abnormal MRI, multiple seizure types, and mental and motor deficits [8,10,37–40]. In validating this questionnaire, it will be important to address the pitfalls faced by the current instruments. Many of the currently available instruments to assess seizure severity such as the Hague Seizure Severity Scale for Children and the E-CHESS are limited [14,15,20,41]. The E-CHESS was developed and validated for use in infants [20]. The Hague, while catering to a broader age range, is fraught with challenges regarding clinical relevance, scaling issues, and responsiveness of the scale in relation to clinical changes [14,15,41]. These difficulties with clinical relevance, scaling issues, and responsiveness of scale in relation to clinical changes are also seen with the QOLCE [12,26,28,29]. Thus, it will be important to assess the responsiveness of the PREQ to change and establish clinical relevance of the scores computed from the PREQ as future studies are carried out to develop the questionnaire further. 5. Conclusion This pilot study is an initial, critical step in the development of the PREQ. The significant correlations between the PREQ subscale scores and the external epilepsy severity and quality of life measures lend preliminary support to our hypothesis that the PREQ is assessing the severity of epilepsy along with other important domains, such as mood, development, and quality of life. A larger prospective study of
this modified PREQ is currently underway to further develop the PREQ and validate it as a quantitative instrument that can be used to reliably assess the clinical status of children with MRE undergoing palliative epilepsy treatment. Acknowledgments The authors thank Larry Snyder, M.D., Ph.D. and Jay F. Piccirillo, M.D. for their help in research design. The authors thank Janice E. BrunstromHernandez, M.D., Michael A. Ciliberto, M.D., Michael J. Noetzel, M.D., John M. Zempel, M.D., Ph.D., Liu-Lin Thio, M.D., Ph.D., Judy Weisenberg, M.D., Michael Wong, M.D., Ph.D., Rohini Coorg, M.D., Bryan McGill, M.D., Ph.D., and Adam Ostendorf, M.D. for their help in patient recruitment and participation in the study. The authors thank Deanna Mercer for her help with Institutional Board Review. This research was funded by the Washington University School of Medicine Fund. Conflict of interest statement None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2014.04.025. References [1] Shinnar S, Pellock JM. Update on the epidemiology and prognosis of pediatric epilepsy. J Child Neurol 2002;17(Suppl. 1):S4–S17. [2] Bailet LL, Turk WR. The impact of childhood epilepsy on neurocognitive and behavioral performance: a prospective longitudinal study. Epilepsia 2000;41(4):426–31. [3] Bjornaes H, et al. The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study. Seizure 2001;10(4):250–9. [4] Callenbach PM, Westendorp RG, Geerts AT, Arts WF, Peeters EA, van Donselaar CA, et al. Mortality risk in children with epilepsy: the Dutch study of epilepsy in childhood. Pediatrics 2001;107(6):1259–63. [5] Breningstall GN. Mortality in pediatric epilepsy. Pediatr Neurol 2001;25(1):9–16. [6] Strzelczyk A, Reese JP, Dodel R, Hamer HM, Callenbach PM. Cost of epilepsy: a systematic review. Pharmacoeconomics 2008;26(6):463–76. [7] O'Dell C, Wheless JW, Cloyd J. The personal and financial impact of repetitive or prolonged seizures on the patient and family. J Child Neurol 2007;22(5 Suppl.): 61S–70S. [8] Seker Yilmaz B, Okuyaz C, Komur M. Predictors of intractable childhood epilepsy. Pediatr Neurol 2013;48(1):52–5. [9] Pressler RM. Vagus nerve stimulation in children with intractable epilepsy. Dev Med Child Neurol 2012;54(9):782–3. [10] Berg AT, Shinnar S, Levy SR, Testa FM, Smith-Rapaport S, Beckerman B. Early development of intractable epilepsy in children: a prospective study. Neurology 2001;56(11):1445–52. [11] Nei M, O'Connor M, Liporace J, Sperling MR, Callenbach PM, Callenbach PM. Refractory generalized seizures: response to corpus callosotomy and vagal nerve stimulation. Epilepsia 2006;47(1):115–22. [12] Sabaz M, Lawson JA, Cairns DR, Duchowny MS, Resnick TJ, Dean PM, et al. The impact of epilepsy surgery on quality of life in children. Neurology 2006;66(4):557–61. [13] Baker GA, Smith DF, Jacoby A, Hayes JA, Chadwick DW. Liverpool Seizure Severity Scale revisited. Seizure 1998;7(3):201–5. [14] Cramer JA. Assessing the severity of seizures and epilepsy: which scales are valid? Curr Opin Neurol 2001;14(2):225–9. [15] Carpay JA, Vermuelen J, Stroink H, Brouwer OF, Peters AC, Aldenkamp AP, et al. Seizure severity in children with epilepsy: a parent-completed scale compared with clinical data. Epilepsia 1997;38(3):346–52. [16] Engel Jr J, E. International League Against. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001;42(6):796–803. [17] Wieser HG, Blume WT, Fish D, Goldensohn E, Hufnagel A, King D, et al. ILAE Commission Report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001;42(2):282–6. [18] Camfield C, Breau L, Camfield P. Impact of pediatric epilepsy on the family: a new scale for clinical and research use. Epilepsia 2001;42(1):104–12. [19] O'Donoghue MF, Duncan JS, Sander JW. The National Hospital Seizure Severity Scale: a further development of the Chalfont Seizure Severity Scale. Epilepsia 1996;37(6): 563–71.
V. Purusothaman et al. / Epilepsy & Behavior 37 (2014) 26–31 [20] Humphrey A, Ploubidis GB, Yates JR, Steinberg T, Bolton PF, Callenbach PM. The Early Childhood Epilepsy Severity Scale (E-Chess). Epilepsy Res 2008;79(2–3):139–45. [21] Speechley KN, Sang X, Levin S, Zou GY, Eliasziw M, Smith M, et al. Assessing severity of epilepsy in children: preliminary evidence of validity and reliability of a singleitem scale. Epilepsy Behav 2008;13(2):337–42. [22] Kanner AM. The complex epilepsy patient: intricacies of assessment and treatment. Epilepsia 2003;44(Suppl. 5):3–8. [23] Pellock JM. Defining the problem: psychiatric and behavioral comorbidity in children and adolescents with epilepsy. Epilepsy Behav 2004;5(Suppl. 3):S3–9. [24] Loring DW, Meador KJ, Lee GP. Determinants of quality of life in epilepsy. Epilepsy Behav 2004;5(6):976–80. [25] Wheless JW. Intractable epilepsy: a survey of patients and caregivers. Epilepsy Behav 2006;8(4):756–64. [26] Titus JB, Lee A, Kasasbeth A, Thio LL, Stephenson J, Steger-May K, et al. Health-related quality of life before and after pediatric epilepsy surgery: the influence of seizure outcome on changes in physical functioning and social functioning. Epilepsy Behav 2013;27(3):477–83. [27] Speechley KN, Ferro MA, Camfield CS, Huang W, Levin SD, Smith ML, et al. Quality of life in children with new-onset epilepsy: a 2-year prospective cohort study. Neurology 2012;79(15):1548–55. [28] Tanriverdi T, Olivier NP, Olivier A. Quality of life after extratemporal epilepsy surgery: a prospective clinical study. Clin Neurol Neurosurg 2008;110(1):30–7. [29] Sabaz M, Lawson JA, Cairns DR, Duchowny MS, Resnick TJ, Dean PM, et al. Validation of the Quality of Life in Childhood Epilepsy Questionnaire in American epilepsy patients. Epilepsy Behav 2003;4(6):680–91. [30] Morita DA, Glauser TA, Modi AC. Development and validation of the Pediatric Epilepsy Side Effects Questionnaire. Neurology 2012;79(12):1252–8.
31
[31] Kothare SV, Wagner J. Quality of life in pediatric epilepsy: a validated questionnaire for side effects of AEDs. Neurology 2012;79(12):1194–5. [32] Wei Y, Oakland T, Algina J. Multigroup confirmatory factor analysis for the adaptive behavior assessment system-II parent form, ages 5–21. Am J Ment Retard 2008;113(3):178–86. [33] Vohr BR, Stephens BE, Higgins RD, Bann CM, Hintz SR, Das A, et al. Are outcomes of extremely preterm infants improving? Impact of Bayley assessment on outcomes. J Pediatr 2012;161(2):222–8 [e3]. [34] Fayers PM, Groenvold M, Hand DJ, Bjordal K. Clinical impact versus factor analysis for quality of life questionnaire construction. J Clin Epidemiol 1998;51(3):285–6. [35] Lautenschlager GJ. A comparison of alternatives to conducting Monte-Carlo analyses for determining parallel analysis criteria. Multivar Behav Res 1989;24(3): 365–95. [36] Kulkarni AV, Rabin D, Drake JM. An instrument to measure the health status in children with hydrocephalus: the Hydrocephalus Outcome Questionnaire. J Neurosurg 2004;101(2 Suppl.):134–40. [37] Ko TS, Holmes GL. EEG and clinical predictors of medically intractable childhood epilepsy. Clin Neurophysiol 1999;110(7):1245–51. [38] Ohtsuka Y, Yoshinaga H, Kobayashi K. Refractory childhood epilepsy and factors related to refractoriness. Epilepsia 2000;41(Suppl. 9):14–7. [39] Chawla S, Aneja S, Kashyap R, Mallika V. Etiology and clinical predictors of intractable epilepsy. Pediatr Neurol 2002;27(3):186–91. [40] Berg AT, Levy SR, Novotny EJ, Shinnar S. Predictors of intractable epilepsy in childhood: a case–control study. Epilepsia 1996;37(1):24–30. [41] Cramer JA, French J. Quantitative assessment of seizure severity for clinical trials: a review of approaches to seizure components. Epilepsia 2001;42(1):119–29.
