Technical variability of the RT3 accelerometer.

PURPOSE To evaluate the technical performance of the RT3 triaxial accelerometer. METHODS Twenty-three RT3 accelerometers were subjected to a specific vibration along each sensitive axis in isolation, using a motorized vibration table that produced frequencies of 2.1, 5.1, and 10.2 Hz, respectively. Data were analyzed for frequency and axis effects and inter- and intra-instrument variability. RESULTS ANOVA showed a frequency by axis interaction (F2.1,36.8 = 19.9, P < 0.001). Post hoc tests revealed the Y axis count to be significantly higher than the X and Z axes counts at 5.1 and 10.2 Hz. There was no difference in counts between axes at 2.1 Hz. Interinstrument coefficients of variation (CV) decreased as frequency increased (21.9 to 26.7% at 2.1 Hz, 6.3 to 9.0% at 5.1 Hz, and 4.2 to 7.2% at 10.2 Hz). The intraclass correlation (ICC) between RT3s was 0.99, regardless of the axis. Intra-instrument CV also decreased as frequency increased (2.1 to 56.2%, 0.3 to 2.5%, and 0.2 to 2.9% at 2.1, 5.1, and 10.2 Hz, respectively. CONCLUSION There were no differences in counts recorded on the X, Y, and Z axes at 2.1 Hz; however, the counts recorded along the Y axis were significantly higher than the counts at the X and Z axes at 5.1 and 10.2 Hz. Due to large coefficients of variation for both inter- and intra-instrument variability at 2.1 Hz, testing the inter- and intra-instrument variability of the accelerometers before use is recommended.