Author ' s response to reviews

1. Re-word ‘in the way they usually do’ to “in everyday clinical practice”. We changed the text as follows (abstract and methodology section): ‘Therapists were instructed to examine the patients according to the methods used in their everyday clinical practice.’ 2. The results section should include p-values. We supplemented the following text to the results section: ‘The differences in portions of elevated plantar pressures between the GS and the ratings per discipline show significant discrepancies (all p-values < .05).’ Background: 3. Re word ‘felt-therapy’ to “padding and strapping therapy.” 4. Re-word ‘shoeware’ to “footwear”. We changed the text as follows: ‘Off-loading strategies could include padding and strapping therapy, foot orthoses, therapeutic footwear or total contact casts.’ 5. Reference(s) required after various techniques such as footprints...... We added the following references: • Kannus, V. P. (1992). "Evaluation of abnormal biomechanics of the foot and ankle in athletes." Br J Sports Med 26(2): 83-9. • Hughes, J., S. Kriss, et al. (1987). "A clinician's view of foot pressure: a comparison of three different methods of measurement." Foot Ankle 7(5): 277-84. • Silvino, N., P. M. Evanski, et al. (1980). "The Harris and Beath footprinting mat: diagnostic validity and clinical use." Clin Orthop(151): 265-9. • Mathieson, I., D. Upton, et al. (1999). "Comparison of footprint parameters calculated from static and dynamic footprints." The Foot 9: 145-199. • Welton, E. A. (1992). "The Harris and Beath footprint: interpretation and clinical value." Foot Ankle 13(8): 462-8. 6. Can the authors please explain the ‘amount’ of plantar pressure? The evidence about 700KPa as the threshold needs to be fully justified. We changed the text as follows: ‘Advanced pressure sensitive devises i.e. platforms and in-shoe devises, provide information about the location as well as the magnitude of plantar peak pressure’ Seven hundred kPa as cut-off criterion for categorizing peak pressure into elevated or non-elevated pressure was based on values which are frequently cited in the literature pertaining diabetes, rheumatoid arthritis and metatarsalgia. Studies in healthy subjects showed that peak pressures were significantly lower than 700 kPa. We are aware that 700 kPa as a risk-threshold is based on weak evidence. We deal with this issue in the discussion section and show the results of lower cut-off points. We changed the text as follows and added the following references: ‘In the literature, peak pressure values similar or higher than 700 kPa are frequently considered as abnormal or elevated. Also, studies in healthy subjects showed that peak pressures were significantly lower than 700 kPa.’ • Cavanagh, P. et al. (2000). "New developments in the biomechanics of the diabetic foot." Diabetes Metab Res Rev 16 Suppl 1: S6-S10. • Armstrong, D. G., (1998). "Is there a critical level of plantar foot pressure to identify patients at risk for neuropathic foot ulceration?" J Foot Ankle Surg 37(4): 303-7. • Peters, E.J., et al. (2001). "Effectiveness of the diabetic foot risk classification system of the International Working Group on the Diabetic Foot." Diabetes Care 24(8): 1442-7. • Frykberg, RG, (1998) "Role of neuropathy and high foot pressures in diabetic foot ulceration. " Diabetes Care 21:1714–1719, (8 –10). • Lavery, L. A., et al. (2003). "Predictive value of foot pressure assessment as part of a populationbased diabetes disease management program." Diabetes Care 26(4): 1069-73. • Malay, D.S., (1996) “Mechanically induced metatarsalgia.” Gait and Posture (4): 198-199. • Otter, S. J., et al. (2004). "Forefoot plantar pressures in rheumatoid arthritis." J Am Podiatr Med Assoc 94(3): 255-60. • van der Leeden, M., et al. (2006). "Forefoot joint damage, pain and disability in rheumatoid arthritis patients with foot complaints: the role of plantar pressure and gait characteristics." Rheumatology 45(4): 465-9. • Hsi, W. L., et al. (2005). "Optimum position of metatarsal pad in metatarsalgia for pressure relief." Am J Phys Med Rehabil 84(7): 514-20. • Hayafune, N., et al. (1999). "Pressure and force distribution characteristics under the normal foot during the push-off phase in gait." The Foot 9: 88–92. • Luger, E. J., et al. (1999). "Patterns of weight distribution under the metatarsal heads." J Bone Joint Surg Br 81(2): p199-202. • Bryant, A. R, et al. (2000). "Normal values of plantar pressure measurements determined using the EMED-SF system." J Am Podiatr Med Assoc 90(6): 295-9. • Cavanagh, P., et al. (1987). "Pressure distribution under symptom-free feet during barefoot standing." Foot Ankle 7(5): 262-76. • Bennett, P., et al. (1993). "Pressure distribution beneath the human foot." J Am Podiatr Med Assoc 83(12): 674-287. • Hennig, E. M., et al. (1994). "Plantar pressure distribution patterns of young school children in comparison to adults." Foot Ankle Int 15(1): 35-40. 7. Sentence about the three patients should be moved into the methodology section. We adopted the suggestion of the reviewer. Methods: 8. The mean of five steps was taken using a platform system. Did the authors eliminate any steps or was it the first five consecutive steps? 9. How were the regions identified? Current clinical evidence suggests ‘masking’ discrete regions of the foot. Bare foot peak pressures were estimated per foot, by calculating the mean over the readings of 5 consecutive measurements. After instruction and practice of the plantar pressure measurement procedure, data of five steps per foot were collected. The procedure for masking was according to the preceding BMC paper, but now applied to bare foot plantar pressure data. We changed the text as follows: ’Bare foot peak pressures were estimated per foot, by calculating the mean over the readings of 5 consecutive measurements. This was done for six discrete regions: big toe (BT) and metatarsal one (mt-1) to five (mt-5) defined through Novel ‘create any mask®’ software and verified through anterior-posterior radiographs.’ • Weijers, R. E., G. H. I. M. Walenkamp, et al. (2003). "The relationship of the position of the metatarsal heads and peak plantar pressure." Foot and Ankle Int 24: 349-353. 10. I cannot comment on the statistical tests. I assume the data was normally distributed to use a mixed-model ANOVA. Please explain the principles of Streiner and Norman. We would kindly suggest chapter 8 ‘Generalizability Theory’ of the excellent book ‘Health Measurement Scales: a practical guide to their development and use’ by Streiner D.L. and Norman G.R.(1995). A summary of the described principles is no alternative for their concise and clear text. Although we think it would not be suitable to integrate the following text in the present paper, we shall attempt to elucidate the principles described by Streiner and Norman: Streiner and Norman’s calculation of the ICC (intraclass correlation coefficient, which can be used as a measure of method agreement: ICC = R), involves the definition of individual patient variation plus the definition of specific patient variation originating from differential methods. The individual patient variation can be estimated by the variance component for random factor patient i.e. Varcomp (P), the specific patient variation originating from differential methods can be estimated by the variance component for the interaction of the random factor patient and the fixed factor method i.e. Varcomp (Pm). The ratio of Varcomp (P) / (Varcomp (P) + Varcomp (Pm)) results in the ICC for method agreement. However, in multivariate Generalizability Theory, the calculation can be more complex because other factors besides ‘P’ and ‘m’, may be involved in the ANOVA design: in the present study ‘therapist’ within discipline, ‘side’ within patient and ‘region’ within foot. Streiner and Norman suggest a strategy to incorporate variance components involving interaction between patient and aforementioned factors in both numerator and denominator of the ratio. Define X as the set of other factors. All variance components involving factor ‘P’ are used in the calculation of the concerned ICC or so-called rho coefficient: Varcomp (P) + Varcomp (Px) R = Varcomp (P) + Varcomp (Px) + Varcomp (Pm) + Varcomp (Pmx) Where P = patient or subject x = other relevant factors, such as discipline, therapist, side (left/right), region (within foot), etc. m = method i.e. Gold standard vs. clinical rating 11. It is interesting to note that the authors decided to accept 0.80 as being acceptable. Previous authors have used a less robust system, where 0.6 is acceptable. For example, Portney and Watkins [2] suggest that an ICC > 0.75 is deemed good; between 0.50-0.75 as moderate and <0.50 represents poor reliability. Bruton [3] suggests that an intra-class correlation of at least 0.6 is to be considered useful. We agree there are no absolute standard values for acceptable reliability using the ICC. The whole context of measurement and type of discipline determines how much reliability is needed. Portney and Watkins indeed state that ‘.75 is indicative of good reliability’, but they continue with the statement that ‘for many clinical measurements reliability, reliability should exceed .90 to ensure reasonable validity’ (p. 565). As we point out in the second paragraph of the discussion, we consider that identification of locations with elevated pressures is an important aspect in foot care decision-making. According to the statement in the method section, we replaced the word sufficient with acceptable. Results: 12. I am unclear of Table 3. Please can explain in more detail the concept of observed proportions of elevated pressure? For example the GS for the big toe is 0.50. What is the SI unit? The figures in table 3 represent dimensionless proportions, which allow to compare the ratio of regions with elevated pressure according to the GS and ratio of regions indicated by therapists. We supplemented the legend o