78 Štajer, Burger, Vidmar / Rehabilitacija - letn. VII, supl. 3 (2008) MEASUREMENTS FOR ORTHOPAEDIC SHOES: A STUDY ON ACCURACY AND AGREEMENT OF PROFESSIONALS AND STUDENTS T. Štajer, H. Burger, G. Vidmar T. Štajer, H. Burger, G. Vidmar Institute for Rehabilitation, Ljubljana, Slovenia Institute for Rehabilitation, Ljubljana, Slovenia analysed. In general, agreement within both groups was very high, whereby technicians were more synchronised among themselves than students. We also found that tech- nicians appropriately determined EU shoe size as two to three sizes larger than the foot length, which is measured by the scanner and was erroneously reported by the students. Therefore, efforts should be made within undergraduate orthotics and prosthetics education to familiarise the students with practical procedures and requirements of orthopaedic measurements. Abstract We aimed at comparing foot measurements of final- year Prosthetics & Orthotics students (ten; having knowledge, but lacking experience), skilled techni- cians (two; lacking formal education on foot anatomy and disorders, but having lots of experience) and the measures provided by a foot scanner. Both feet of four healthy adult volunteers were measured. Agreement with scanner and within-group agreement were statistically INTRODUCTION Many people, especially elderly, have foot problems. These impact their activities, such as walking and consequently also participation (1). Foot disorders can affect all the more proximal joints of the lower limb (2). There are several reasons for foot problems, such as diabetes, rheumatoid arthritis and different injuries, but also inappropriate shoes. Inappropriate shoes cause foot ulcers in up to 20% of patients with diabetes (3). Over 70% of elderly patients admitted to a general rehabili- tation unit have been found to have inappropriate shoes (3). Many patients with foot problems get orthopaedic shoes. It is important that the measurement is performed by a skilled person who also has appropriate knowledge of foot anatomy and deformities. The aim of the present study was to compare foot measure- ments of final-year Prosthetics & Orthotics (P&O) students (having knowledge, but lacking experience), skilled techni- cians (lacking formal education on foot anatomy and disor- ders, but having 10 years of experience) and the measures provided by a scanner. METHODS AND SUBJECTS Methods The measurements were performed by two skilled techni- cians from the Institute for Rehabilitation, Republic of Slovenia, Ljubljana, and ten P&O students from the Uni- versity College of Health Studies. They were all instructed to perform the measurements for the purpose of producing orthopaedic shoes. Additionally, the subjects’ feet were measured by a foot scanner (UCS Inc., Vrhnika, Slovenia), which only measured the dimensions A, B, C and F (see below and figure 1). Six dimensions were measured (Figure 1): A. circumference at metatarsal heads (in cm); B. circumference at midfoot (in cm); C. circumference at hindfoot (in cm); D. ankle circumference at malleolar level (in cm); E. shank circumference at 15 cm (in cm); F. EU shoe size. Figure 1: Foot measures (left) and sample output from the foot scanner (right) Mean deviation (MD) was used to assess agreement of the measurers with the scanner measurements. Coefficient of variation (i.e., standard deviation divided by the mean, expressed in %) was used to assess agreement within the two groups of measurers. Intraclass correlation (ICC, using 79 absolute agreement definition with two-way random effects model) was used to assess overall agreement among the measurers (4). Bland-Altman plots (5) were used to further explore agreement and accuracy of the measurers. Subjects Subjects Four adult volunteers were measured (three male and one female), aged 28-59 years. All were healthy and none had any foot impairment or deformity. Both feet of each subject were measured. RESULTS Descriptive statistics summarising the measurements, agreement with scanner and agreement among each group of measurers are reported in Table 1. Mean CV over all dimensions for the technicians was 1.1% and 1.2% for left and right foot, respectively (and 1.1% over both feet), while for students it was 2.4% and 2.2% (2.3 % overall). Selected bland-Altman plots comparing the technicians and the stu- dents against the scanner (Figure 2) illustrate good agree- ment regarding circumference at hindfoot (top panels) and downward bias of the students and the scanner with respect to the technicians (bottom panels). Table 1: Descriptive statistics summarising the measurements, agreement with scanner and agreement among each group of measurers (MD = mean deviation; CV = coefficient of variation; NA=not available). r e n n a c S s t n e d u t S s n a i c i n h c e T Left Foot Right Foot Left Foot Right Foot Left Right Subject Dimension Mean (Range) MD CV Mean (Range) MD CV Mean (Range) MD CV Mean (Range) MD CV Foot Foot A 27.0 (27.0-27.0) 1.10 0% 27.3 (27.0-27.5) 0.85 1% 24.7 (23.0-28.5) -1.25 6% 24.4 (23.5-26.0) -2.05 3% 25.9 26.4 B 25.8 (25.5-26.0) -1.85 1% 26.0 (26.0-26.0) -1.40 0% 26.3 (25.0-27.5) -1.35 3% 26.2 (25.5-27.5) -1.23 2% 27.6 27.4 C 37.5 (38.0-37.0) 0.20 2% 37.0 (37.5-36.5) 0.20 2% 37.1 (36.0-38.0) -0.25 2% 36.7 (36.0-37.0) -0.15 1% 37.3 36.8 D 24.5 (24.0-25.0) NA 3% 24.5 (24.0-25.0) NA 3% 28.2 (26.0-30.0) NA 4% 28.1 (26.0-30.0) NA 4% NA NA E 25.0 (24.5-25.5) NA 3% 25.0 (24.5-25.5) NA 3% 25.5 (25.0-27.0) NA 3% 25.6 (25.0-26.5) NA 3% NA NA 1 F 45.0 (45.0-45.0) 2.50 0% 45.0 (45.0-45.0) 3.00 0% 43.2 (42.0-44.0) 0.65 1% 42.7 (42.0-43.5) 0.65 1% 42.5 42.0 A 23.0 (23.0-23.0) -0.40 0% 23.3 (23.0-23.5) -0.15 2% 22.9 (22.0-24.5) -0.55 3% 22.9 (22.0-23.5) -0.50 2% 23.4 23.4 B 22.5 (22.0-23.0) -1.40 3% 22.5 (22.0-23.0) -0.70 3% 23.2 (22.0-24.5) -0.70 3% 23.3 (22.5-24.5) 0.05 3% 23.9 23.2 C 32.8 (33.0-32.5) 1.05 1% 32.8 (33.0-32.5) 2.25 1% 32.0 (31.0-34.0) 0.30 3% 31.9 (31.0-33.0) 1.37 2% 31.7 31.7 D 22.5 (22.0-23.0) NA 3% 21.8 (21.5-22.0) NA 2% 23.6 (22.0-25.0) NA 4% 23.7 (22.0-25.0) NA 4% NA NA E 23.0 (23.0-23.0) NA 0% 22.8 (23.0-22.5) NA 2% 23.2 (22.5-24.0) NA 2% 23.1 (22.0-24.0) NA 3% NA NA 2 F 39.3 (39.0-39.5) 2.25 1% 39.3 (39.0-39.5) 1.75 1% 38.0 (37.5-39.0) 1.00 1% 38.0 (37.5-38.5) 0.50 1% 37.0 37.5 A 28.0 (28.0-28.0) 0.50 0% 27.3 (27.5-27.0) 0.05 1% 26.2 (25.0-27.5) -1.30 3% 26.2 (24.5-28.0) -1.05 4% 27.5 27.2 B 28.8 (28.5-29.0) -1.65 1% 28.8 (28.5-29.0) -0.85 1% 28.7 (27.0-30.0) -1.75 3% 29.2 (28.0-30.5) -0.45 3% 30.4 29.6 C 39.5 (40.0-39.0) 0.10 2% 39.0 (39.5-38.5) 0.10 2% 39.3 (39.0-40.0) -0.15 1% 38.9 (38.0-40.0) 0.00 2% 39.4 38.9 D 27.0 (27.0-27.0) NA 0% 27.0 (27.0-27.0) NA 0% 29.4 (28.0-31.0) NA 3% 29.6 (28.5-31.0) NA 2% NA NA E 28.0 (28.0-28.0) NA 0% 28.3 (28.5-28.0) NA 1% 29.1 (27.0-39.0) NA 12% 28.1 (27.0-29.0) NA 2% NA NA 3 F 45.3 (45.5-45.0) 3.25 1% 45.3 (45.5-45.0) 2.75 1% 42.7 (41.5-44.0) 0.70 2% 42.6 (41.0-43.5) 0.05 2% 42.0 42.5 A 29.5 (29.0-30.0) 0.70 2% 28.3 (28.5-28.0) 0.25 1% 27.6 (27.0-29.0) -1.20 2% 27.3 (26.0-29.0) 0.75 4% 28.8 28.0 B 28.8 (28.5-29.0) -1.95 1% 28.8 (28.5-29.0) -1.35 1% 29.2 (28.5-30.5) -1.50 2% 29.4 (29.0-30.0) -0.75 1% 30.7 30.1 C 39.3 (39.5-39.0) -1.25 1% 39.3 (39.5-39.0) -0.25 1% 38.8 (38.0-39.0) -1.75 1% 38.4 (37.0-40.0) -1.10 2% 40.5 39.5 D 28.8 (28.0-29.5) NA 4% 28.3 (27.5-29.0) NA 4% 29.9 (28.0-32.0) NA 4% 29.8 (28.0-33.0) NA 5% NA NA E 28.8 (28.5-29.0) NA 1% 28.0 (28.0-28.0) NA 0% 30.1 (28.5-39.5) NA 11% 28.1 (27.0-29.0) NA 3% NA NA 4 F 45.8 (45.5-46.0) 2.25 1% 45.8 (45.5-46.0) 2.25 1% 42.9 (42.0-44.0) -0.60 2% 43.3 (42.5-44.5) -0.25 2% 43.5 43.5 A 26.9 (23.0-30.0) 0.48 0.6% 26.5 (23.0-28.5) 0.25 1.3% 25.3 (22.0-29.0) -1.08 3.8% 25.2 (22.0-29.0) -1.09 3.2% 26.4 26.3 B 26.4 (22.0-29.0) -1.71 1.7% 26.5 (22.0-29.0) -1.08 1.4% 26.8 (22.0-30.5) -1.33 2.6% 27.0 (22.5-30.5) -0.60 2.2% 28.2 27.6 C 37.3 (32.5-40.0) 0.03 1.4% 37.0 (32.5-39.5) 0.58 1.4% 36.8 (31.0-40.0) -0.46 1.6% 36.5 (31.0-40.0) 0.03 1.8% 37.2 36.4 D 25.7 (22.0-29.5) NA 2.4% 25.4 (21.5-29.0) NA 2.1% 27.8 (22.0-32.0) NA 3.9% 27.8 (22.0-33.0) NA 4.1% NA NA E 26.2 (23.0-29.0) NA 1.0% 26.0 (22.5-28.5) NA 1.4% 26.9 (22.5-39.5) NA 7.1% 26.2 (22.0-29.0) NA 2.7% NA NA Overall Mean F 43.8 (39.0-46.0) 2.56 0.6% 43.8 (39.0-46.0) 2.44 0.6% 41.7 (37.5-44.0) 0.44 1.7% 41.6 (37.5-44.5) 0.24 1.3% 41.3 41.4 Štajer, Burger, Vidmar / Rehabilitacija - letn. VII, supl. 3 (2008) 80 In general, agreement within both groups was very high (ICC computed over all dimensions was 0.996 and 0.975 for technicians and students, respectively, and 0.972 for the pooled sample). Figure 2: Bland-Altman plots for comparing technicians (left panels) and students (right panels) with scanner (thin black line = mean difference, thick grey lines = mean dif- ference ± 2SD; circles = female, diamonds = male subjects; open symbols = left, filled symbols = right foot) DISCUSSION It is encouraging that in general, agreement within both groups was very high. As expected, the technicians were more synchronised among themselves than were the stu- dents. Students’ measurements tended to follow the foot scanner "mechanically" on all dimensions, which is not desirable for the purposes of producing orthopaedic shoes, especially regarding shoe size. The technicians appropriately determined EU shoe size as two to three sizes larger than C Scanner-Mean (Technicians) -3 -2,5 -2 -1,5 -1 -0,5 0 0,5 1 1,5 2 29 31 33 35 37 39 41 Scanner-Mean (Students) -2 -1,5 -1 -0,5 0 0,5 1 1,5 2 2,5 29 31 33 35 37 39 41 [Scanner + Mean (Technicians)] / 2 ► [Scanner + Mean (Students)] / 2 ► F Scanner-Mean (Technicians) -4 -3,5 -3 -2,5 -2 -1,5 -1 -0,5 0 37 38 39 40 41 42 43 44 45 Scanner-Mean (Students) -2 -1,5 -1 -0,5 0 0,5 1 37 39 41 43 45 the foot length, which is measured by the scanner and was also erroneously reported by the students. CONCLUSION In undergraduate orthotics and prosthetics education, atten- tion should be paid on practical experience. In particular, the students should be familiarised with the procedures and requirements of actual orthopaedic shoe measurements. R References eferences: 1. Novak P, Burger H, Marinček Č, Meh D. Influence of foot pain on walking ability of diabetic patients. J Rehab Med, 2004; 36 (6): 249-252. 2. Redford JB, Basmajian JV , Trautman P . Orthotics: clini- cal practice and rehabilitation technology. New York: Churchill Livingstone 1995: 337. 3. Burns LS, Leese GP, McMurdo MET. Older people and ill fitting shoes. Postgrad Med J 2002;78 (920): 344- 346. 4. McGraw K, Wong S. Forming inferences about some intraclass correlation coefficients. Psych Meth 1996; 1 (1): 30-46. 5. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measure- ment. Lancet 1986; i: 307-310. Štajer, Burger, Vidmar / Rehabilitacija - letn. VII, supl. 3 (2008)