Psychosocial and quality-of-life (QOL) outcomes in adult patients with implantable cardioverter-defibrillators (ICDs) are well studied. Minimal research exists regarding pediatric adjustment, despite a potentially more challenging adjustment process. The purpose of the present study was to examine psychosocial and QOL functioning of pediatric ICD patients from patient and parent self-reports. Children and primary caregiver dyads from several university hospitals were analyzed using the PedsQL, the Device Severity Index, the ICD and Avoidance Survey, and demographic information. Sixty children (25 female, 35 male) were enrolled. The present pediatric sample reported lower psychosocial and physical QOL scores than healthy children’s normative scores. In comparison with a sample of chronically ill children, pediatric ICD patients reported lower physical QOL. Parent-observed QOL reports revealed lower psychosocial and physical QOL than parent-observed healthy norms and lower psychosocial and physical QOL than chronically ill norms. There were no QOL differences by ICD shocks or medical severity. Female patients reported lower psychosocial, physical, and cardiac QOL scores. Children reported better QOL than parent observations on psychosocial and physical health. Finally, 84.7% of children reported avoidance behaviors since ICD implantation, with female children avoiding places more than male children. In conclusion, pediatric ICD patients are comparable to children with other chronic illnesses with the exception of lower physical QOL. Similar to adult samples, female children reported poorer QOL and were more likely to engage in avoidance behaviors. Patients indicated better QOL perceptions than their parents’ reports. ICD discharges and medical severity did not negatively affect QOL.
In the present study, we examined quality of life (QOL) in pediatric patients with implantable cardioverter-defibrillators (ICDs). The study addressed the following aims: (1) to make normative comparisons in QOL between healthy and chronically ill norms, (2) to identify differences in QOL between male and female pediatric ICD patients, (3) to identify differences in QOL between patients with positive shock histories and those who had not experienced shocks, and (4) to compare patient self-reports and parent-observed reports of QOL.
Methods
Participants were pediatric ICD patients and their respective parents or guardians during routine electrophysiology visits at 1 of 6 major United States medical centers. After institutional review board approval at each participating center, potential participants and their parents or guardians were approached by medical staff members while attending routine cardiology or electrophysiology appointments. For study eligibility, patients were required to meet the following inclusion criteria: (1) past ICD implantation, (2) age 8 to 18 years, (3) ability to read and complete questionnaires in English, and (4) accompanied by a consenting parent or guardian aged ≥18 years. Patients were excluded from consideration in this study if they had ever been diagnosed with mental retardation or psychotic disorders. None of the potential participants approached met the criteria for exclusion. After obtaining informed consent and assent, the pediatric ICD patients and their parents or guardians were given separate self-report psychosocial assessment packets. The packets took approximately 20 minutes to complete and were completed in the clinic waiting rooms or returned at the next clinic visit. Staff members were available to help the patients and their parents or guardians for clarification.
Each parent or guardian completed a demographic questionnaire to obtain the following information about the patient and parent or guardian: (1) age, (2) gender, (3) race or ethnicity, and (4) highest grade completed. Parent or guardian marital status and average annual household income were also obtained. Statistical analyses were performed to evaluate the proposed hypotheses for this research project using SPSS version 17.0 (SPSS, Inc., Chicago, Illinois).
The Defibrillator Severity Index is a cardiologist-designed rating scale that allows physicians and medical staff members to classify ICD patients in a standardized format on the basis of the severity of their underlying disease. The following domains are covered in the 8-item index: (1) cardiovascular disease, (2) surgical treatment, (3) ICD indications, (4) ventricular arrhythmias, (5) current arrhythmia medications, (6) appropriate ICD discharges, (7) co-morbid medical conditions, and (8) activity limitations. Each item is rated by degree of severity, with higher ratings indicating increased severity. This measure has shown acceptable cardiologist interrater reliability (Pearson’s r = 0.90, κ = 0.51, p <0.01).
The PedsQL Pediatric Quality of Life Inventory–Child Versions was used to assess pediatric QOL. The PedsQL versions that were used in this study include the core module, measuring generic QOL in the last month, with separate versions for children aged 8 to 12 years and teens aged 13 to 18 years. This 23-item module assesses 4 scales (Physical, 8 items; Emotional, 5 items; Social, 5 items; and School, 5 items) that generate 3 summary scores: (1) Total Scale Score (23 items), (2) Physical Health Summary Score (8 items), and (3) Psychosocial Health Summary Score (15 items). All items are statements with which the child is asked to rate his or her agreement on a 5-item, Likert-type scale ranging from “never” to “almost always” and include statements such as “It is hard to pay attention in class” and “I worry about what will happen to me.” Cronbach’s α values for reliability for the Total Scale Score (α = 0.88), Physical Health Summary Score (α = 0.80), and Psychosocial Health Summary Scale (α = 0.83) are all within acceptable limits for group comparisons. In terms of validity, the PedsQL differentiates not only between healthy children and acutely or chronically ill children but also between levels of disease severity within a given medical condition. The presence of a pediatric chronic health condition was provided by parents’ reports of their children as having asthma, diabetes, attention-deficit hyperactivity disorder, depression, or “other.” The chronically ill sample consisted of 574 children ranging in age from 2 to 16 years who in the previous month were significantly more likely to have missed more school days, spent more days sick in bed or too ill to play, and needed someone to care for them because of their illnesses.
In addition, the PedsQL 3.0 Cardiac Module was used to measure cardiac-specific QOL, with versions for children aged 8 to 12 years and teens aged 13 to 18 years. Statements from this disease-specific 27-item module include statements such as “My lips turn blue when I run,” and “I don’t like other people to see my scars.” This module has 5 scales assessing (1) symptoms (7 items), (2) perceived physical appearance (3 items), (3) treatment anxiety (4 items), (4) cognitive problems (5 items), and (5) communication (3 items).
The PedsQL Pediatric Quality of Life Inventory–Parent Versions 3 5 was used to assess pediatric QOL, as answered by parents about their children. The PedsQL parent versions used in this study include the core module, measuring generic QOL, with separate versions for parents to answer for their 8- to 12-year-old children or for their 13- to 18-year-old teens. Like the child version, this version is a 23-item measure of 4 core scales that generate 3 summary scores. The parent versions of this measure have shown similar reliability to the child versions (Total Scale Score α = 0.90, Physical Health Summary Score α = 0.88, and Psychosocial Health Summary Scale α = 0.88). The statements are slightly modified from the child version and ask parents or guardians to rate how much of a problem their children have had in each of the measured domains in the past month.
The parent proxy version of the PedsQL 3.0 Cardiac Module was used to measure parent-perceived cardiac-specific QOL, with parent versions for children aged 8 to 12 years and teens aged 13 to 18 years. This 27-item measure asks parents to rate their agreement with statements about their children’s heart-related functioning on a 5-point, Likert-type scale ranging from “never” to “almost always.” The parent proxy version of this module has the same 6 scales as the child version. Psychometric testing of the parent proxy version of this module has demonstrated satisfactory internal consistency reliability across all scales, ranging from r = 0.81 to r = 0.95.
The ICD and Avoidance Survey is a questionnaire that includes questions regarding whether and to what extent respondents have begun avoiding places, objects, and activities since receiving an ICD. The instrument was initially trialed with a group of elderly volunteers, but there are no current validity or reliability estimates. Still, this measure has high face validity and is written in simple and clear language.
Results
A total of 60 patients whose ages ranged from 8 to 18 years were enrolled. Fifty-eight percent of the study sample (n = 35) was male. There was no significant difference in age between male and female patients, t (57) = 1.70, p = 0.095. Of the male patients (n = 35), 62.9% (n = 22) were identified as Caucasian, 20% (n = 7) identified themselves as Hispanic or Latino, 11.4% (n = 4) self-identified as African American, 2.9% (n = 1) identified as other, and 2.9% (n = 1) did not identify themselves in this category. Of the female patients (n = 25), 77.3% (n = 17) were identified as Caucasian, 18.2% (n = 4) identified themselves as Hispanic or Latino, 9.1% (n = 2) self-identified as African American, 4.5% (n = 1) identified as Asian or Pacific Islander, and 4.5% (n = 1) did not identify themselves in this category. Participating legal guardians were primarily mothers (75%), with fathers (15%), grandparents (3.3%), other legal guardians (1.7%), and nonresponders (5%) constituting a smaller percentage of the population. Sixty-eight percent of the adults in the sample reported being married or remarried, and the median annual family income ranged from $30,000 to $59,999. Roughly half of the sample (48.3%) had experienced ≥1 previous ICD shock.
We expected significant differences between healthy norms compared with the present sample. There was also the expectation that our sample would be similar to normative data among pediatric patients with other chronic illnesses. Several 1-sample Student’s t tests were conducted to make normative comparisons with the present sample of pediatric ICD patients. Normative data were accessed from Varni et al’s reliability and validity study on the PedsQL 4.0. The present pediatric ICD sample reported significantly lower psychosocial health and physical health than healthy norms ( Table 1 ). Scores also reflected significantly lower psychosocial health than pediatric patients with other chronic illnesses, but physical health scores were not significantly different ( Table 2 ).
| PedsQL Scale | Present Sample | Healthy Norms | t | p Value | ||
|---|---|---|---|---|---|---|
| n | Mean ± SD | n | Mean ± SD | |||
| Child | ||||||
| Total | 55 | 70.7 ± 17.7 | 5,079 | 83.9 ± 12.5 | 5.60 | <0.001 |
| Psychosocial | 57 | 69.7 ± 18.2 | 5,070 | 81.8 ± 14.0 | 5.08 | <0.001 |
| Physical | 58 | 72.6 ± 19.4 | 5,070 | 87.8 ± 13.1 | 5.91 | <0.001 |
| Parent | ||||||
| Total | 57 | 67.3 ± 16.2 | 8,713 | 82.3 ± 15.6 | 6.97 | <0.001 |
| Psychosocial | 57 | 66.5 ± 17.1 | 8,714 | 81.2 ± 15.3 | 6.49 | <0.001 |
| Physical | 58 | 68.4 ± 19.0 | 8,696 | 84.1 ± 19.7 | 6.30 | <0.001 |
| PedsQL Scale | Present Sample | Chronic Health Condition | t | p Value | ||
|---|---|---|---|---|---|---|
| n | Mean ± SD | n | Mean ± SD | |||
| Child | ||||||
| Total | 55 | 70.7 ± 17.7 | 574 | 74.2 ± 15.4 | 1.48 | 0.14 |
| Psychosocial | 57 | 69.7 ± 18.2 | 573 | 71.3 ± 17.1 | 2.68 | 0.01 |
| Physical | 58 | 72.6 ± 19.4 | 574 | 79.5 ± 17.1 | 0.67 | 0.50 |
| Parent | ||||||
| Total | 57 | 67.3 ± 16.2 | 831 | 73.1 ± 16.5 | 2.71 | 0.01 |
| Psychosocial | 57 | 66.5 ± 17.1 | 830 | 71.0 ± 17.3 | 1.99 | 0.51 |
| Physical | 58 | 68.4 ± 19.0 | 830 | 77.0 ± 20.2 | 3.46 | <0.01 |
In terms of parent-observed QOL, reports revealed significantly lower psychosocial health and physical health than parent ratings of healthy norms ( Table 1 ). Parent-observed QOL scores also reflected significantly lower physical health and psychosocial health scores than parent ratings of pediatric patients with other chronic illnesses ( Table 2 ).
To investigate any differences in QOL between shocked and nonshocked pediatric ICD patients, an omnibus 1-way analysis of variance was used, with shock (yes or no) used as the fixed variable and the PedsQL summary scales (Physical Health Summary Scale, Psychosocial Summary Scale, and Cardiac Scale) as dependent variables. On the basis of adult ICD research, it was expected that shocked patients would not have significantly lower QOL scores than nonshocked patients, at least until the experience of ≥5 shocks. The sample did not contain a sufficient number of participants who had experienced multiple shocks, so analyses could not be conducted to test this hypothesis. Shock did not have a significant effect on any of the QOL variables per child report and parent report.
To identify if statistically controlling for ICD discharge was necessary when comparing QOL between genders, we examined potential differences in the experience of shock between genders. An independent-samples Student’s t test was conducted, comparing male and female patients’ reports of shock. Statistical analyses indicated no significant difference in the experience of an ICD discharge, t (58) = 1.624, p = 0.11, between male (mean 1.43 ± 0.502) and female (mean 1.64 ± 0.490) patients in this sample, so there was no need to control for ICD discharge when examining gender differences in QOL.
To determine effects of medical severity on QOL, an omnibus 1-way analysis of variance was used, with Defibrillator Severity Index scores (low or high) used as the fixed variable and the PedsQL summary scales (Physical Health Summary Scale, Psychosocial Summary Scale, and Cardiac Scale) as dependent variables. Statistical analyses revealed that pediatric patients with greater medical severity reported significantly lower QOL than those with lower medical severity in psychosocial health, F(1, 38) = 7.14, p = 0.011, and physical health, F(1, 36) = 7.50, p = 0.01.
To identify if statistically controlling for medical severity was necessary when comparing QOL between genders, we examined potential differences in medical severity between genders in this sample. An independent-samples Student’s t test was conducted, comparing male and female patients’ total scores on the Defibrillator Severity Index. The male patients in our sample had a mean score of 7.89 ± 2.81. The female patients had a mean score of 8.78 ± 3.83. Statistical analyses indicated no significant difference (p = 0.427) in medical severity between genders in this sample and did not warrant statistical control when examining gender differences in QOL.
Because there were no differences in shock status or medical severity by gender in this sample, we investigated differences in QOL between male and female pediatric ICD patients by conducting an omnibus 1-way analysis of variance, with gender used as the fixed variable and the PedsQL summary scales (Physical Health Summary Scale, Psychosocial Summary Scale, and Cardiac Scale) as dependent variables. On the basis of the extant adult ICD research, it was expected that female pediatric ICD patients would report lower levels of QOL than their male counterparts. Statistical analyses revealed that female pediatric patients reported significantly lower QOL than male patients in psychosocial health, physical health, and cardiac health ( Table 3 ). Scores of pediatric QOL by parent proxy failed to demonstrate significant differences between female and male patients ( Table 3 ). Interestingly, female patients in the sample were significantly more likely to avoid places than men, t (54) = 2.239, p = 0.029.
