In order to gain a full understanding of the reliability of a measure, it has been recommended that systematic change in mean, absolute and relative consistency are all reported [19]. To the authors knowledge, this was the first study to determine the reliability of a PFD device during two dynamic plantar flexion protocols. Previously, researchers have primarily focused on isometric plantar flexion assessments. The main findings of this research were: 1) in terms of intra-session reliability, all peak force variables were found to have acceptable reliability for both testing protocols (change in mean ≤ 3.0%, CV ≤ 3.3%, ICC > 0.94), while peak RFD development had poor to moderate reliability (change in mean ≤ 11.1%, CV ≤ 20.7%. ICC > 0.74); 2) with regards to inter-session reliability, once again, peak force variables were found to have acceptable reliability (change in mean ≤ 3.8%, CV ≤ 5.3%, ICC > 0.89), while peak RFD had poor to moderate reliability (change in mean ≤ 16.2%, CV ≤ 27.0, ICC > 0.47); 3) there appeared no significant differences between left and right for the dorsi-flexed position, however peak RFD for the neutral position was significantly different (p < 0.05, 8.35%, ES = -0.82): and, 4) in terms of protocol differences, values were significantly higher for peak force (p < 0.01, 6.44–8.10%, ES = 1.05–1.75) and peak RFD (p < 0.01, 34.91–66.22%, ES = 1.30–1.52) during the neutral position.
In terms of intra-session reliability, both protocols had acceptable absolute (Neutral: 2.1–3.3%, Dorsi: 1.6–2.9%) and relative consistency (Neutral: 0.95–0.98, Dorsi: 0.97–0.99) for peak force variables. With regards to peak RFD measures, absolute and relative consistency appeared better on the left leg (Neutral: CV = 5.7–8.5%, ICC = 0.87–0.94; Dorsi: CV = 8.1–12.4%, ICC = 0.88–0.96) compared to the right leg (Neutral: CV = 13.7–20.7%, ICC = 0.75–0.92; Dorsi: CV = 13.5–15.2%, ICC = 0.93–0.94) for both protocols. Force plates have a similar sampling frequency (1000 Hz) to the PFD device used in this study (1200 Hz) and have been reported to have excellent intra-session reliability for single leg standing isometric plantar flexion (CV = 3.2%, ICC = 0.93–0.97) for peak force variables (relative and absolute) [16]. With regards to RFD, there is a paucity of reliability research that has utilized this force-time metric, nonetheless RFD measures have been reported to be typically less reliable than maximum force, especially during the early phase of contraction [21]. With this in mind, further refining of the protocol is required to improve the consistency of this measure.
Similar results were found for between session reliability for peak force during both neutral and dorsi-flexed protocols (CV = 3.0–3.9%, ICC = 0.93–0.96; and CV = 2.8–5.3%, ICC = 0.90–0.97, respectively). Once again, peak RFD was poor to moderate at best for both neutral and dorsi-flexed protocols (CV = 10.8–18.5%, ICC = 0.48–0.81; and CV = 23.9–27.0%, ICC = 0.49–0.76, respectively). As with intra-session reliability, previous research has only focused on peak force metrics. Mattiussi and colleagues [16], reported similar reliability to that of the current study for single leg plantar flexion isometric peak force (CV = 3.0–4.9%, ICC = 0.96–0.98). Given the intra-session results of peak RFD, it is not surprising the poor intersession reliability was observed. Though the authors cannot make direct comparisons to similar protocols, these results are similar to those reported by Juneau and colleagues [22], using similar strain gauge technology during an isometric knee extension (CV = 16.4–24.7%, ICC = 0.50–0.78).
Regarding interlimb asymmetry, there was no significant differences between left and right legs for peak force (p > 0.05, 1.10%, ES = 0.18) and peak RFD (p > 0.05, 11.74%, ES = 0.23) in the dorsi-flexed position. Similar results were seen for peak force in the neutral position (p > 0.05, 0.57%, ES = -0.15), however peak RFD was significantly different between legs, the right leg values greater (p < 0.05, 8.35%, ES = -0.82). This finding is likely because all but one participant reported their right leg as their dominant side.
In terms of protocol differences, all metrics were significantly higher in the neutral position, compared to the dorsi-flexed position (PF = p < 0.01, 6.44–8.10%, ES = 1.05–1.75 and peak RFD = p < 0.01, 34.91–66.22%, ES = 1.30–1.52). This was no surprise, as previous researchers reported on the effects of joint angle on rate of torque development during a prone isometric explosive contraction, reporting higher peak torque and rate of torque development in a neutral position, compared to a dorsi-flexed position, due to a more optimal length tension relationship [23]. These findings highlight the importance of being strict with set up procedures, as small changes in ankle angle can significantly affect the results. Additionally, practitioners may wish to test at shorter muscle lengths during early-stage rehabilitation, and then progressing their athlete/patient to longer muscle lengths during late-stage rehabilitation. Finally, the reader needs to be cognizant that the testing position results are not interchangeable and needs to be wary when comparing plantar flexor results.