Epilepsia · 2007 · 55 citations · 20 references
MedicineNeurologyEpilepsyEpileptologySleep MedicineElectroencephalographyElectroencephalography In EpilepsySeizure ClassificationSleep AssessmentSleep-wake DisordersEpilepsy DiagnosisInterobserver Reliability AssessmentKappa StatisticNocturnal Frontal Lobe EpilepsySeizuresSleepTonic SeizuresVideo Electroencephalography Classification
BACKGROUND: Nocturnal frontal lobe seizures (NFLS) show one or all of the following semeiological patterns: (1) paroxysmal arousals (PA: brief and sudden recurrent motor paroxysmal behavior); (2) hyperkinetic seizures (HS: motor attacks with complex dyskinetic features); (3) asymmetric bilateral tonic seizures (ATS: motor attacks with dystonic features); (4) epileptic nocturnal wanderings (ENW: stereotyped, prolonged ambulatory behavior). OBJECTIVE: To estimate the interobserver reliability (IR) of video-recording diagnosis in patients with suspected NFLS among sleep medicine experts, epileptologists, and trainees in sleep medicine. METHODS: Sixty-six patients with suspected NFLS were included. All underwent nocturnal video-polysomnographic recording. Six doctors (three experts and three trainees) independently classified each case as "NFLS ascertained" (according to the above specified subtypes: PA, HS, ATS, ENW) or "NFLS excluded". IR was calculated by means of Kappa statistics, and interpreted according to the standard classification (0.0-0.20 = slight agreement; 0.21-0.40 = fair; 0.41-0.60 = moderate; 0.61-0.80 = substantial; 0.81-1.00 = almost perfect). RESULTS: The observed raw agreement ranged from 63% to 79% between each pair of raters; the IR ranged from "moderate" (kappa = 0.50) to "substantial" (kappa = 0.72). A major source of variance was the disagreement in distinguishing between PA and nonepileptic arousals, without differences in the level of agreement between experts and trainees. CONCLUSIONS: Among sleep experts and trainees, IR of diagnosis of NFLS, based on videotaped observation of sleep phenomena, is not satisfactory. Explicit video-polysomnographic criteria for the classification of paroxysmal sleep motor phenomena are needed.
20
The Measurement of Observer Agreement for Categorical Data
J. Richard Landis, Gary G. Koch · Biometrics · 1977 · 78.9K citations · Full text
Statistical Methods for Rates and Proportions.
B. S. Everitt, J. L. Fleiss · Biometrics · 1981 · 9.1K citations