Capturing What Matters: Patient-reported LGI1-ANTibody Encephalitis Outcome RatiNg Scale (LANTERN)
Mark Kelly1, Barbara Wagner2, Bryan Ceronie2, Christine Strippel2, Ann Yee Lin2, Adam Handel2, John Soltys1, Sophie Binks2, Philip Powell3, Sarosh Irani1
1Neurology, Mayo Clinic, 2Oxford Autoimmune Neurology Group, University of Oxford, 3Sheffield Centre for Health and Related Research (SCHARR), University of Sheffield
Objective:
To design the first patient reported outcome measure for LGI1-antibody encephalitis (LGI1-Ab-E).
Background:
LGI1-Ab-E is a common form of autoimmune encephalitis where most patients demonstrate 'good' clinician-rated outcomes. However, more targeted questionnaires reveal numerous debilitating symptoms for many years. To better quantify these persistent features, we designed the LGI1-Antibody Encephalitis Rating (LANTERN) scale, a quantified, disease-specific patient-reported outcome measure (PROM), adhering to FDA guidelines.
Design/Methods:
A participant-driven mixed-methods approach to develop a clinically valid questionnaire over three stages: (1) Item generation through semi-structured interviews; (2) Repeated cognitive debriefing rounds to advance comprehensibility, relevance and comprehensiveness; (3) Psychometric survey to condense the most sensitive and valid questions. Analyses incorporated sensitivity testing with multiple internal and external validations.
Results:
From 73 items across six domains (Stage 1; n = 18), a questionnaire assessing the frequency and severity of 43 symptoms (80 questions), plus nine activities of daily living (ADL), was developed through cognitive debriefing (Stage 2; n = 15). This 89-question survey was completed (Stage 3; n = 66 patients and 32 relatives) and distilled, using exploratory factor analyses, to a three-factor symptom-burden questionnaire comprising 41 questions (19 symptoms and 6 ADL), separated into physical, cognitive/behavioural and ADL domains. These factors demonstrated strong internal reliability (Cronbach alpha: 0.85-0.91), correlations with relative-completed questionnaires (R = 0.73-0.85; p < 0.001), good-to-excellent intraclass re-testing correlations (0.81-0.98; n = 19) and strong associations with numerous predefined external measures.
Conclusions:
LANTERN represents a PROM for LGI1-Ab-E, with initial content, structural and construct validity and test-retest reliability. It can be used as a reliable, tailored, efficient and sensitive method to establish symptom burden in people with LGI1-Ab-E, both in clinical practice and trials.
Generative AI Usage
No, did not use generative AI in the drafting or editing in this abstract.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.