Rationale:
“Children with epilepsy” is a heterogeneous population that includes individuals ranging from neonates to adolescents, neurotypical to developmentally delayed, and “otherwise healthy” to medically complex. Measuring quality of life in this population is challenging because many existing instruments have floor effects that impede measurement in certain sub-populations. The Pediatric Epilepsy Learning Healthcare System Quality of Life (PELHS-QOL-2) may overcome these floor effects due to its sole focus on seizures and anti-seizure medication side effects.
Method:
We obtained data from a cross-sectional study conducted to establish the construct validity of the PELHS-QOL-2, a novel two-item health-related quality of life (HRQOL) instrument for children with epilepsy. Demographics, clinical data, and caregiver responses to the comparator instruments were collected. We identified a group of individuals (“QOLCE-55 Floor”) for whom 15 or more responses to the QOLCE-55 were marked as “Not Applicable” or left blank. We first characterized how the “QOLCE-55 Floor” population differed from their non-floor counterparts, and then we examined whether the PELHS-QOL-2 was still associated with the other comparator instruments for the “QOLCE-55 Floor” population. Qualitatively, two reviewers independently reviewed caregiver feedback in order to identify general themes, create subthemes, and populate each subtheme with a representative quote.
Results:
Compared to children with more complete QOLCE-55 scores, children in the “QOLCE-55 Floor” population tended to be younger (7.00 [2.00-12.0] vs 11.0 [7.00-14.5]), located in inpatient settings (60.4% vs 26.6%), and receiving care at Cornell (60.4% vs 23.4%) at the time of data collection (p < 0.001). The “QOLCE-55 Floor” children were more likely to have poor quality of life per the PELHS-QOL-Seizures (2.96 [1.22] vs 3.55 [0.90]), G-QOLCE (2.00 [2.00-3.60] vs 2.00 [1.00-3.00]), and GASE (3.00 [2.00-5.00] vs 6.00 [4.00-7.00]), in addition to severe developmental delay (68.8% vs 29.9%), complex comorbidities (39.6% vs 14.0%), and more recent seizure activity (52.1% vs 24.3%) (p < 0.01). There were moderate correlations between the PELHS-QOL-Seizures and GASE (rs = 0.57, p < 0.001, n = 46) and G-QOLCE (rs = -0.35, p = 0.017, n = 45), and between the PELHS-QOL-Medications and PESQ (rs = -0.41, p = 0.018 , n = 33), in the “QOLCE-55 Floor” population (Figure). Qualitatively, some caregivers expressed an inability to answer several QOLCE-55 questions because their children were too young or had too many comorbidities (Table).
Conclusion:
The QOLCE-55 both quantitatively and qualitatively demonstrates floor effects due to its limited applicability to younger children with higher comorbidity burden. The PELHS-QOL-2 exhibits construct validity for the “QOLCE-55 Floor” population, indicating that PELHS-QOL-2 may be valuable for a broader group of children with epilepsy.
Funding:
:No funding was received in support of this abstract.
FIGURES
Figure 1