Variables: General Results
Variables: General Results
the sensor, without magnitude great enough to cause discomfort to the operator (given the rigid mounting of the sensor mat),
operators were instructed to produce a "moderate" amount of force and were allowed to practice on the system before data
collection was begun. Because the force durations of manually applied forces28 can be almost two orders of magnitude larger
than those applied by instrument,26 force-recording equipment with high sampling rates is not required. This permits use of
multiplexed sensor matrices, which can characterize changes in contact area known to occur with manual force applications.14
The multiple-sensor mat used in this experiment was calibrated for a maximal pressure of 120 psi, and data were collected by the
manufacturer's proprietary system at a maximal rate of 1000 Hz (Sensor Products Inc.).
Variables
From the load cell and sensor mat data, two primary variables were collected: peak-to-peak force magnitude (N) and force
duration (millisecond). Force magnitude was calculated after any preapplication force was subtracted, with the difference of force
occurring over the ascending limb of the force-time plot. Force duration was computed by counting the number of data samples
occurring in the ascending limb, then dividing this figure by the collection rate.
Variation Analysis. For each primary variable
(force and force duration), mean value and SD were calculated for all 10 trials generated from each unique combination of
instrument and operator (16 instrument combinations and 4 manual combinations). As a general measure of variability in primary
variables and to place data on a common reference scale, the SD of each instrument/operator combination was expressed as a
percentage of its associated mean value. In addition, Levene's test was used as a measure of intraoperator variability wherein a p
value greater than .05 described insignificant between-operator differences in within-operator variation for a given instrument.
Absolute Value Comparison. To determine whether different instrument operators generated different magnitudes of absolute force and
force duration, we performed multiple comparisons on the primary variables for each operator pairing (six possible) for each
instrument and for the manual technique. If Levene's test was insignificant (homogeneous within operator variation), the Tukey
post hoc analysis was performed; if Levene's test was significant, the Tamhane post hoc analysis was performed.
RESULTS
General Results
Manual applications of force were generally greater in force magnitude and force duration than were those delivered by
instrument (see Table 3-1 and Figure 3-1, A and B). The mean force for all manual applications was 264 N, and mean force
duration was 145 milliseconds. For all instrument applications, the average force was 171 N, and average force duration was
0.963 milliseconds.
Variation Analysis
Although greater in magnitude, manual forces also displayed greater deviation from the mean value. On average, the SD for all
manual applications represented 16% of the mean applied force and 23% of the mean force duration (see Table 3-1, Figure 3-1,
Table 3-2, and Figure 3-2, A and B). For all instrument applications, the SD represented 4% of the mean applied force and 5% of
the mean force duration (see Tables 3-1 and 3-2; see Figures 3-1 and 3-2). Results for Levene's test pertaining to intraoperator
variability are described in Table 3-3. For the force variable, intraoperator variation for all instruments and for the manual
technique was found not to differ between operators.
For force duration, the Impulse instrument and the manual technique were shown to have intraoperator variations that differed
significantly between operators (p .01 and p .004, respectively).
= =