INTERLABORATORY VARIABILITY OF ROTATIONAL CHAIR TEST-RESULTS

Test-retest reliability of rotational chair testing for a single facil ity has previously been examined by others. The actual data analysis m ethods, however, have received far less attention. The variety of both hardware and software currently used theoretically may affect the res ults for a given subject tested at different facilities. The purposes of this study were, first, to quantify the amount of variability in th e analysis of identical raw data files at multiple rotational chair te sting facilities by using automated analysis; second, to evaluate the effect of operator intervention on the analysis; and third, to identif y possible sources of variability. Raw data were collected from 10 nor mal subjects at 0.05 Hz and 0.5 Hz (50 degrees per second peak velocit y). Diskettes containing raw electro-oculogram data files were then di stributed to eight participating laboratories for analysis by two meth ods: (I) using automated analysis algorithms and (2) using the same al gorithms but allowing operator intervention into the analysis. Respons e parameters calculated were gain and phase (re: velocity). The SD of gain values per subject for automated analysis ranged from 0.01 to 0.3 2 gain units and of phase values from 0.4 to 13.7 degrees. For analysi s with operator intervention, the SD of gain values ranged from 0.02 t o 0.10 gain units and of phase values from 0.4 to 4.4 degrees. The dif ference between automated analysis and analysis with operator interven tion was significant for gain calculations (p < 0.02) but not for phas e calculations (p > 0.05). This study demonstrates significant variabi lity in automated analysis of rotational chair raw data for gain and p hase. Operator intervention into the analysis significantly reduces va riability for gain but not for phase.