Abstracts

Rationale: Lennox-Gastaut Syndrome is a severe and devastating form of childhood epilepsy. The frequent occurrence of multiple types of seizures combined with developmental delay and behavioural disturbance can have a significant impact on a child’s quality of life (QOL). Utilities represent preferences for health states and are necessary to estimate quality adjusted life years (QALYs). Ranging from 1 (full health) to 0 (dead), utilities can be a useful tool for health care providers and decision makers to evaluate the true benefit of an intervention. Preferences can be measured directly from patients or can be elicited from caregivers, clinical experts or members of the general public. When assessing the cost effectiveness of interventions within a publicly funded health care system, it is generally considered that utilities should reflect societal values. There is an absence of data on public preferences for health states associated with LGS. Therefore, a study was designed to capture utility values for LGS health states. Methods: Four health state (HS) descriptions of LGS outcomes and 5 HS descriptions of common adverse events of anti-epileptic treatments for LGS were developed following medical literature review and extensive consultation with clinical experts. HSs were defined by tonic-atonic (drop attack) seizure frequency (SF). For LGS outcomes, the anchor state (HS-1) for uncontrolled disease was described by frequency of 21-28 seizures per week; HS-2 a reduction of < 50% in SF following adjunctive treatment; HS-3 a reduction of >=50% and <75% in SF and HS-4 a >=75% reduction in SF following adjunctive treatment. Time trade off interviews (TTO) were conducted with 119 members of the general public of whom 48% were caregivers/parents of children aged 4 to 18. A secondary analysis involved participants rating each health state on a visual analogue scale (VAS) and completing the EQ-5D questionnaire. Results: The mean utility score for HS-1, HS-2, H-S3 and HS-4 was 0.393, 0.461, 0.605 and 0.699 respectively using TTO method; 0.02, 0.414, 0.556 and 0.677 respectively using VAS scale; 0.02, 0.100, 0.500 and 0.596 respectively using EQ-5D questionnaires. Differences between all LGS HSs and the anchor state were significant (p<0.0001) except for HS-2 using EQ-5D. The preferences were reasonably consistent across age groups and gender. Caregivers/parents gave slightly higher scores for the health states only via TTO method. For adverse events the disutility score was 0.108 for concentration problems, 0.135 for weight loss, 0.174 for somnolence, 0.190 for rash and 0.193 for Nausea/Vomiting.