Navicular drop measurement in people with rheumatoid arthritis: interrater and intrarater reliability.

Patients with rheumatoid arthritis (RA) frequently develop foot pain, deformity, and difficulty with ambulation. (1-4) Subtalar and midtarsal joint hypermobility is often cited as a primary impairment leading to excessive foot pronation and subsequent functional limitations. (1,4-8) Although it is clear that excessive pronation may lead to several painful foot conditions, it is less clear at what point the rear-foot and midfoot motion becomes abnormal or "excessive." To investigate this relationship, the clinician requires measurement techniques that yield reliable data and valid inferences from the data.

Navicular drop (ND) measurement is gaining popularity with clinicians and researchers for quantifying midfoot mobility and may be a valuable examination technique for patients with RA. Several investigators (9-13) have suggested that ND measurement may be the most valid and reliable static clinical measure of foot pronation currently available to clinicians. Navicular drop measurement is defined as the difference in height of the most prominent aspect of the navicular tuberosity when the subtalar joint is placed in "neutral" (STJN) as compared with when the foot is positioned in a relaxed standing foot posture (RSFP) (Fig. 1). (14,15) Therefore, the measurement is used to quantify midtarsal joint pronation or "flattening" of the medial longitudinal arch during standing. Criterion validity of static navicular height (NH) measurements has been established by radiographic methods, (9) and NH measurements have been shown to be an indicator of navicular bone movement during gait in people without foot deformity. (12) In addition, dynamic navicular bone movement has been shown to correlate with dynamic rear-foot motion. (12) Thus, in subjects without foot pathology, ND measurements appear to be a valid indicator of midfoot and rear-foot pronation under static and dynamic conditions.

Some investigators (16-21) have suggested that larger ND values are associated with musculoskeletal injury and muscular weakness. Navicular drop measurement has been shown to discriminate between females with and without anterior cruciate ligament injuries, (16,17,19,21) with the females with injuries yielding a greater magnitude of ND. Bennett et al (18) found that a relationship existed between larger ND values and development of medial tibial stress syndrome in runners. Snook (20) compared a group of 12 people without foot injuries with [greater than or equal to] 13 mm ND and a matched control group of 12 people with an average of only 7 mm ND. The investigator measured isokinetic peak torque at 30[degrees]/s and found that the subjects with larger ND values had decreased concentric plantar-flexor torque. Navicular drop measurements also have recently been used by researchers for classifying potential subjects as pronators or nonpronators (22-25) and have been used as the primary dependent variable for investigations of controlling midfoot pronation via athletic taping. (26-28)

Intrarater and interrater reliability of ND measurements have been investigated in subjects without foot and ankle impairments (13,14,16,29) (Tab. 1). Brody (15) first described ND measurement and reported values under 10 mm as normal and values over 15 mm as abnormal, but he provided no data to support his observations and did not investigate reliability. Mueller et al (29) reported good intrarater reliability and refined the ND measurement technique by adding an individualized template for each subject to control for variation in foot position between trials. Picciano et al (30) first investigated interrater reliability of ND measurements and found poor reliability, with an intraclass correlation coefficient (ICC [1,1]) of.57. In this study, they did not use foot position templates, and both testers were students who were inexperienced with ND measurement. Sell et al (14) found good intrarater and interrater reliability. Other authors (13,14,16,29) have reported interrater or intrarater reliability of NH measurements used to calculate ND values (Tab. 1).

Because of foot deformities and joint hypermobility associated with RA, a large degree of variability may exist when taking measurements related to the bony morphology. However, we found only one article (a case report (31)) that described the use of ND measures in people with RA. There are no reports of reliability of ND measurements in people with RA.

The primary purpose of this study was to investigate interrater and intrarater reliability of ND measurements in people with RA taken by clinicians with varied experience. We also wanted to investigate the potential for error associated with skin markings used for NH measurements and to compare ND values in people with RA with published normative values. In addition, we were interested in comparing the traditional bilateral stance ND measurement technique with a more recently proposed single-limb stance (SLS) technique.. (18,32) Based on our clinical experience making ND measurements in people with RA, we believed ND values would be greater than the referenced normal values of 6 to 8 mm (Tab. 1). We also noticed that the position of the navicular bone during the RSFP portion of the measurement technique may drop inferior to the skin mark. If this happens, then the validity of the ND measurement would be questionable and would underrepresent midfoot pronation. This would require modification of the measurement technique to include repalpation and remarking of the navicular bone after the patient is instructed to stand in RSFP.

We hypothesized that interrater reliability would be poor to moderate and that intrarater reliability would be moderate to high. Our null hypotheses were that the ND values in patients with RA would not be different from normative values, as reported in the literature, and that no difference would exist between NH measurement in RSFP when using the original skin marking versus repalpating the bony landmark and making a new mark. Our final null hypothesis was that SLS ND values would be no different than those taken from bilateral standing.

Method

Study Design

A repeated-measures design was used in this study in which 4 different NH measurements were repeated 3 times on each foot by each of the 3 examiners during a morning session and then all measurements were repeated during an afternoon session on the same day. The NH measurements constituted a repeated variable, and the examiners constituted an independent, nonrepeated variable.

Subjects

Ten women (20 feet) who were diagnosed with RA participated in this study. All subjects signed an informed consent statement and were recruited at random from a list of patients who were treated at an outpatient rehabilitation clinic of an acute care hospital during the previous 10 years. Inclusion criteria consisted of being age 18 years or older, having the ability to climb 5 steps (to get onto a raised platform), having no previous foot surgery for the past 12 months, having ankle dorsiflexion range of motion greater than 0 degrees from a knee-extended position, and having enough midfoot and rear-foot mobility to be placed in an STJN position. Subjects completed demographic and outcomes questionnaires. Subject demographics were as follows: mean age of 55.4 years (SD=11.4), mean duration of RA of 12.7 years (SD=10.4), and mean body mass index of 28.5 (SD=5.1). Additional data collected included the foot function index (FFI), the American College of Rheumatology (ACR) functional classification, (33) foot pain, and history of previous intervention with foot orthoses and prescription footwear. These additional data were collected for the purpose of describing our subjects' foot pain and functional abilities (Tab. 2). The FFI has been shown to have good testretest reliability (ICC=.87) for use in people with RA. Construct validity was established with factor analysis, resulting in significant Pearson correlations for all except 2 items. (34) The instrument includes a series of foot-specific, self-report questions and uses visual analog scales divided into 3 subscales: pain, disability, and activity limitation. Scores range from 0 to 100 for each subscale and the total score, with higher scores indicating greater impairment. Our subjects reported that currently foot pain did not interfere with their lifestyle or activity level, which was substantiated by the low mean score of 10/100 on the activity limitation subscale and the relatively low mean total FFI score of 24/100 (Tab. 2). Furthermore, all subjects were identified as class I on the ACR functional classification scale (33) (Tabs. 2 and 3).

Examiners

Three examiners made all NH measurements. Examiners 1 and 2 were physical therapists with 14 (JAS) and 4 (GCG) years experience managing patients with foot and ankle disorders and prior clinical use of ND measurement, and examiner 3 was a physical therapist student (JMP) with no experience managing foot and ankle disorders. Similar to other studies, (14,29,32) the examiners took part in four 1-hour practice sessions to refine the measurement technique prior to data collection. In session 1, the examiners placed volunteers without foot or ankle disorders into an STJN position from a bilateral stance position. This was accomplished by palpating the talonavicular joint for congruency as described by Sell et al. (14) During the second session, examiners alternately palpated and discussed navicular tuberosity landmarks. The examiners completed all NH measurements with a digital height gauge * for 3 volunteer co-workers with divergent foot profiles during sessions 3 and 4.

Measurements and Instrumentation

Measurements were: (1) NH from STJN; (2) NH from an RSFP; (3) NH from an RSFP with repalpation and placement of a new skin marking on the navicular tuberosity, if needed; and (4) NH from SLS with identical repalpation technique. Measurements made in order (1-4) were repeated 3 times, with old skin markings removed and new skin marks placed for all trials. The entire procedure was repeated during an afternoon session separated by a minimum of 2 hours for a total of 6 trials of NH measurements 1 through 4 for each foot by each examiner. Subjects were instructed to stand and bear weight equally for measurements 1 through 3 and to shift their full weight to one limb for measurement 4. Contralateral toe-touch weight bearing was allowed for maintaining balance during measurement 4. From NH measurements, the following ND values were obtained: ND1=NH2-NH1 (as reported in the literature (14,15)), ND2=NH3-NH1 (skin mark error represented by difference between ND1 and ND2), and ND3=NH4-NH1 (SLS). The subjects were instructed to stand, and the examiners placed a mark on the skin at the most inferior border (ledge) of the most prominent medial aspect of the navicular tuberosity. The skin marks were removed prior to the entry of each subsequent examiner. The order of examiners making NH measurements was randomized for each session. Examiners were never in the same room together and did not communicate with one another until data collection was completed. A digital height gauge (Fig. 2) was zeroed to the supporting surface height prior to each measurement. Paper and tape were used to conceal the liquid crystal display from view of the examiners (Fig. 2). A monitor, independent of the study, viewed and recorded all digital measurements on a data collection sheet. A foot position template was made by one examiner (JMP) for each subject prior to data collection. Subjects were asked to march in place for a few seconds and relax into a standing position with a "comfortable" toe-out angle. This foot position was then traced onto butcher's paper, creating the template (Fig. 2). Each subject stood on their template for all measurements in order to standardize foot position for all examiners.

Data Analysis

Descriptive statistics were used to calculate means and standard deviations for all NH and ND measurements. Intraclass correlation coefficients (2,1 and 2,k [k=3 trials]) were used to calculate interrater and intrarater reliability. Paired t tests were used to compare differences between ND measurements. Linear correlations were calculated between ND1 and ND2. Correlations also were calculated between the delta (ND2-ND1) and ND1 to determine whether the magnitude of skin error increased in subjects with larger ND1 values. The standard error of measurement (SEM) was calculated for each NH measurement.

Results

The means and standard deviations for all 4 NH measurements and all 3 ND calculations are presented in Table 4. For interrater reliability, ICCs (2,1) ranged from .67 to .92 and ICCs (2,k) ranged from .85 to .97 (Tab. 5). Intrarater reliability was better than interrater reliability, with ICCs (2,1) ranging from .73 to .95 and ICCs (2,k) ranging from .90 to .98 (Tab. 6). The means of ND1 and ND2 were statistically different at P=.001 to .007 for each examiner and session. In addition, a linear correlation between ND1 and ND2 was strong (r=.995, P<.001). These results, coupled with good interrater reliability for the NH2 and NH3 measurements, indicate that there was a highly reliable difference between the standard NH2 measurement and our suggested NH3 measurement that accounts for skin marking error. Specifically, ND2 values were 1.59 mm and 1.28 mm greater than ND1 values for sessions 1 and 2, respectively, when we repalpated the navicular bone during the RSFP NH3 measurement. Thus, the navicular bone dropped lower than the original skin marking (skin mark error) after our subjects relaxed in RSFP. In order to determine whether subjects with larger ND values had more skin mark error than subjects with smaller ND values, we calculated a linear correlation between the ND delta (ND2-ND1) and ND1 for each subject. This correlation was statistically significant but weak (r =.448, P<.05) (Fig. 3).

Discussion

Our results indicate that NH measurements and ND calculations can be reliable for patients with RA. Consistent with previous findings, (14) both intrarater and interrater reliability were high for ND measurements taken by clinicians with experience managing foot and ankle disorders. In addition, we showed that a physical therapist student with minimal practice and training (4 hours) also can obtain measurements with high intrarater and interrater reliability, which is in contrast to other research. (30) We selected statistical tools (ICC models 2,1 and 2,k) that allow for generalization of our results to all practicing physical therapists. We found that intrarater reliability was higher than interrater reliability, as expected and consistent with previous studies. (14,30)

Interrater Reliability

When we analyzed data from only the first trial of each session with no averaging, we found fair-to-excellent interrater reliability (ICC [2,1]=.67-.90) for session 1 and good-to-excellent reliability (ICC [2,1] =.83-.92) for session 2. We cannot attribute this difference between sessions to an examiner "learning curve" rationale because sessions were performed on the same day for each subject, but subjects were seen on different days. When the average of 3 trials was used for analysis using the ICC 2,k model where k=3 trials, ICCs for interrater reliability of all measurements were excellent, improving to .85 to .96 for session 1 and to .94 to .97 for session 2. Interrater reliability (ICC) of NH1 measurements from the STJN position improved from .77 to .91 for session 1 and from .85 to .94 for session 2 by averaging 3 measurements. The STJN measurement will likely always have the lowest reliability compared with other NH measurements because it requires the examiner to place the patient into a defined weight-bearing position (STJN) via palpation, and then the patient must maintain this position while the NH is measured.

It is plausible that the subjects may have attained some level of "learning" with respect to maintaining this position, leading to an improved NH1 measurement reliability from session 1 (ICC=.77) to session 2 (ICC=.85) for nonaveraged data. If this is true, then clinicians may want to perform 1 or 2 NH practice measurements prior to collecting clinical data to allow patients to fully understand what is expected of them. Therefore, if clinicians or researchers want to attain excellent interrater ND measurement reliability data, they should make each measurement 3 times and then average the results after the patient has practiced maintaining the required foot positions once or twice. To further refine the measurement technique, four 1-hour practice sessions should be sufficient to yield high interrater reliability data (based on our study and that of Sell et al (14)). Our rationale for including practice sessions was to ensure that the 3 examiners were using the same technique and to become familiar with using a digital height gauge. Clinicians who are primarily interested in intrarater data reliability for their particular clinical setting may not require practice sessions.

One additional explanation for why our measurements yielded higher reliability data during the second session may relate to our subjects' concern about possible foot pain during testing. Seven of our 10 subjects had complained of foot pain during the week of testing and may have had reservations about pain escalation while standing on a hard wooden platform with no external foot support while measurements were made. Patients with RA frequently do not tolerate this activity for long periods of time. In order to minimize this possibility, we encouraged all subjects to sit down after each trial of NH measurements (1-4), so they were not required to stand for longer than 1 minute at a time. Perhaps after the first session was completed and subjects perceived no pain, they were less anxious and better able to concentrate on holding the STJN position during the second session.

Intrarater Reliability

Intrarater reliability for all NH measurements was good to excellent (ICC=.73-.95) when using only the first trial and excellent (ICC=.90-.98) when 3 measurements were averaged. We believe that ICC values above .75 are indicative of good reliability. However, this is only a guideline, and practitioners must use their best judgment when reviewing studies to determine what is acceptable for their particular clinical or research setting. (35)

Skin Marking Error

We hypothesized that using the same skin mark for measures NH1 and NH2 could lead to error if the navicular bone drops inferior to the skin mark. Our results proved this to be true because we found a highly reliable difference between the NH2 measurements and the NH3 repalpation measurements for all trials, sessions, and examiners. Given that ND is a relatively small measure of displacement, with normal values between 6 and 8 mm, a 1.5-mm difference could represent an error of 19% to 25%. For this reason, we suggest that clinicians who use skin markings during ND measurement palpate and mark the skin over the navicular bone each time the bone is allowed to move. The increased time for this minor method change was minimal in our experience.

Navicular Drop Values of People With RA Compared With Normative Data

Our subjects' mean ND value of 8.3 mm was slightly higher than the reported normal range of 6 to 8 mm (Tab. 1). When error associated with skin movement was accounted for, the ND values increased to 9.8 mm. We expected mean values to be higher than 10 mm because people with RA commonly have midfoot hypermobility and excessive pronation characterized by medial longitudinal arch flattening. We must first suggest that our sample size (n=10 [20 feet]) was too small for our data to be considered as "normative" data for people with RA. Second, in looking at our ND data range (1.8-20.0 mm), we noticed that 50% of our subjects (10 feet) had ND values less than 7 mm, which is considered normal,13A4,16,29 and 60% of those subjects had an RA disease duration of 15 to 38 years. Patients with long-standing RA can develop stiffness of rear-foot and midfoot joints if the disease begins to "burn out" or go into remission. (36) This could parry explain our finding of mean ND values lower than we expected for our subjects. However, given that larger ND values ([greater than or equal to]13-15 mm) are thought to be related to an increased risk for injury or weakness, (17,20) it follows that our subjects with relatively low ND values ([bar.x]=9.8 mm with skin error correction and 17/20 feet <13 mm) may have less risk for injury and less pain and disability. This claim is supported by our subjects' low FFI scores and unanimous class I ACR functional class status. In addition, all of our subjects had previously received custom-made foot orthoses and prescription footwear intended to partially limit excessive midfoot pronation, which has been shown to alleviate foot pain and improve gait speed and functional mobility in people with RA. (31,37-40)

Navicular Drop From a Single-Limb Standing Position

Two groups of authors (18,32) have investigated modified ND methods that replace the bilateral stance RSFP measure with a unilateral stance RSFP measure. They based their rationale on the work of McPoil and Cornwall, (41) who showed that the static angle of the rear foot in single-limb standing may serve as an indicator of the degree of maximum pronation during the first 60% of the walking cycle in people without foot and ankle disorders. Although both groups of authors (18,32) utilized SLS NH measures, they did not include a traditional bilateral stance measure for comparison. McPoil and Cornwall, (41) however, noted that the rear foot pronated 3.5 degrees farther when subjects attained an SLS position as compared with a bilateral RSF position. Interestingly, this was not the case for our subjects with RA as measured by ND. There was no statistical difference between the mean values for ND2 (RSFP), which we believe accounted for skin marking error, and ND3 (SLS), which presumably represents end-range midtarsal joint pronation. This could imply that our subjects were already at end-range pronation when standing in bilateral RSFP.

Conclusions

Our results suggest that ND measurements taken with the methods described have very good intrarater and interrater reliability, especially when 3 measurements are averaged. To our knowledge, this work represents the first time ND measurement has been studied in patients with known foot and ankle impairments. Our subjects with RA exhibited a mean ND measurement values of 8.3 mm, only slightly higher than the published normal values of 6 to 8 mm reported by authors with comparable methods and reliability data. (13,14,16,29) However, when we added a separate RSFP NH measurement with repalpation of the navicular tuberosity to compare with the traditional RSFP NH measurement, we found that skin movement led to a systematic error. When this error was accounted for, our subjects' ND values increased to 9.8 mm (a difference of 1.5 mm). For this reason, we suggest that clinicians palpate and remark the navicular bone for each NH measurement or avoid the use of skin markings during NO measurement. (16,20,29)

The ND values in our subjects with RA did not increase when they moved from a bilateral stance RSF position to an SLS RSF position. This finding may indicate that our subjects attained maximum midfoot pronation during bilateral resting standing foot posture instead of SLS, as might be expected based on the work of McPoil and Cornwall. (41) Future research is encouraged to determine normative ND values in people with RA with a larger sample size and to assess relationships between ND values and common foot and ankle impairments.

* Sciencescope Inc, 5751 Schaefer Ave, Chino, CA 91710.

References

(1) Cracchiolo A III. Rheumatoid arthritis: hindfoot disease. Clin Orthop. 1997; (340):58-68.

(2) Dimonte P, Light H. Pathomechanics, gait deviations, and treatment of the rheumatoid foot: a clinical report. Phys Ther. 1982;62: 1148-1156.

(3) Kitaoka HB. Rheumatoid hindfoot. Orthop Clin North Am. 1989;20: 593-604.

(4) Vainio K. The rheumatoid foot: a clinical study with pathological and roentgenological comments. Ann Chir Gynaecol Fenn. 1956;45:1-107.

(5) Gerber LH, Hunt GC. Evaluation and management of the rheumatoid foot. Bull N Y Acad Med. 1985;61:359-368.

(6) Hunt GC, Fromherz WA, Gerber LH, Hurwitz SR. Hindfoot pain treated by a leg-hindfoot orthosis: a case report. Phys Ther. 1987;67: 1384-1388.

(7) Thompson F, Mann R. Arthritides. In: Mann R, Coughlin M, eds. Surgery of the Foot and Ankle. St Louis, Mo: Mosby Inc; 1993:615-672.

(8) Vidigal E, Jacoby RK, Dixon AS, et al. The foot in chronic rheumatoid arthritis. Ann Rheum Dis. 1975;34:292-297.

(9) Williams DS, McClay IS. Measurements used to characterize the foot and the medial longitudinal arch: reliability and validity. Phys Ther. 2000;80:864-871.

(10) Saltzman CL, Nawoczenski DA, Talbot KD. Measurement of the medial longitudinal arch. Arch Phys Med Rehabil. 1995;76:45-49.

(11) Menz HB. Alternative techniques for the clinical assessment of foot pronation. J Am Podiatr Med Assoc. 1998;88:119-129.

(12) Cornwall MW, McPoil TG. Relative movement of the navicular bone during normal walking. Foot Ankle Int. 1999;20:507-512.

(13) McPoil TG, Cornwall MW. The relationship between static lower extremity measurements and rearfoot motion during walking. J Orthop Sports Phys Ther. 1996;24:309-314.

(14) Sell KE, Verity TM, Worrell TW, et al. Two measurement techniques for assessing subtalar joint position: a reliability study. J Orthop Sports Phys Ther. 1994;19:162-167.

(15) Brody DM. Techniques in the evaluation and treatment of the injured runner. Orthop Clin North Am. 1982;13:541-558.

(16) Allen MK, Glasoe WM. Metrecom measurement of navicular drop in subjects with anterior cruciate ligament injury. J Athl Train. 2000; 35:403-406.

(17) Beckett ME, Massie DL, Bowers KD, Stoll DA. Incidence of hyperpronation in the ACL injured knee: a clinical perspective. J Athl Train. 1992;27:58-61.

(18) Bennett JE, Reinking MF, Pluemer B, et al. Factors contributing to the development of medial tibial stress syndrome in high school runners. J Orthop Sports Phys Ther. 2001;31:504-510.

(19) Loudon JK, Jenkins W, Loudon KL. The relationship between static posture and ACL injury in female athletes. J Orthop Sports Phys Ther. 1996;24:91-97.

(20) Snook AG. The relationship between excessive pronation as measured by navicular drop and isokinetic strength of the ankle musculature. Foot Ankle Int. 2001 ;22:234 -240.

(21) Woodford-Rogers B, Cyphert L, Denegar CR. Risk factors for anterior cruciate ligament injury in high school and college athletes. J Athl Train. 1994;29:343-346.

(22) Lange B, Chipchase L, Evans A. The effect of low-dye taping on plantar pressures, during gait, in subjects with navicular drop exceeding 10 mm. J Orthop Sports Phys Ther. 2004;34:201-209.

(23) Rose HM, Shultz SJ, Arnold BL, et al. Acute orthotic intervention does not affect muscular response times and activation patterns at the knee. J Athl Train. 2002;37:133-140.

(24) Hargrave MD, Carcia CR, Gansneder BM, Shuhz SJ. Subtalar pronation does not influence impact forces or rate of loading during a single-leg landing. J Athl Train. 2003;38:18-23.

(25) Glasoe WM, Allen MK, Kepros T, et al. Dorsal first ray mobility in women athletes with a history of stress fracture of the second or third metatarsal. J Orthop Sports Phys Ther. 2002;32:560-565; discussion 565-567.

(26) Holmes CF, Wilcox D, Fletcher JP. Effect of a modified, low-dye medial longitudinal arch taping procedure on the subtalar joint neutral position before and after light exercise. J Orthop Sports Phys Ther. 2002;32:194-201.

(27) Vicenzino B, Feilding J, Howard R, et al. An investigation of the anti-pronation effect of two taping methods after application and exercise. Gait Posture. 1997;5:1-5.

(28) Vicenzino B, Griffiths SR, Griffiths LA, Hadley A. Effect of antipronation tape and temporary orthotic on vertical navicular height before and after exercise. J Orthop Sports Phys Ther. 2000;30:333-339.

(29) Mueller MJ, Host JV, Norton BJ. Navicular drop as a composite measure of excessive pronation. J Am Podiatr Med Assoc. 1993;83: 198-202.

(30) Picciano AM, Rowlands MS, Worrell T. Reliability of open and closed kinetic chain subtalarjoint neutral positions and navicular drop test. J Orthop Sports Phys Ther. 1993;18:553-558.

(31) Shrader JA, Siegel KL. Nonoperative management of functional hallux limitus in a patient with rheumatoid arthritis. Phys Ther. 2003;83:831-843.

(32) Vinicombe A, Raspovic A, Menz HB. Reliability of navicular displacement measurement as a clinical indicator of foot posture. J Am Podiatr Med Assoc. 2001;91:262-268.

(33) Hochberg MC, Chang RW, Dwosh I, et al. The American College of Rheumatology 1991 revised criteria for the classification of global functional status in rheumatoid arthritis. Arthritis Rheum. 1992;35: 498-502.

(34) Budiman-Mak E, Conrad KJ, Roach KE. The foot function index: a measure of foot pain and disability. J Clin Epidemiol. 1991;44:561-570.

(35) Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. 2nd ed. Upper Saddle River, NJ: Prentice-Hall Inc; 2000:565.

(36) Clayton ML, Ries MD. Functional hallux rigidus in the rheumatoid foot. Clin Orthop. 1991:233-238.

(37) Fransen M, Edmonds J. Off-the-shelf orthopedic footwear for people with rheumatoid arthritis. Arthritis Care Res. 1997;10:250-256.

(38) Shrader JA, Siegel KL. Postsurgical hindfoot deformity of a patient with rheumatoid arthritis treated with custom-made foot orthoses and shoe modifications. Phys Ther. 1997;77:296-305.

(39) Woodburn J, Barker S, Helliwell PS. A randomized controlled trial of foot orthoses in rheumatoid arthritis. J Rheumatol. 2002;29: 1377-1383.

(40) Woodburn J, Helliwell PS, Barker S. Changes in 3D joint kinematics support the continuous use of orthoses in the management of painful rearfoot deformity in rheumatoid arthritis. J Rheumatol. 2003;30: 2356-2364.

(41) McPoil TG, Cornwall MW. Relationship between three static angles of the rearfoot and the pattern of rearfoot motion during walking. J Orthop Sports Phys Ther. 1996;23:370-375.

JA Shrader, PT, CPed, is Senior Clinical Specialist and Foot Clinic Coordinator, Physical Therapy Section, Department of Rehabilitation Medicine, Mark O. Hatfield Clinical Research Center, National Institutes of Health, Department of Health and Human Services, 10 Center Dr, CRC Rm 1-1469, Bethesda, MD 20892-1604 (USA) ([e-mail]). Address all correspondence to Mr Shrader.

JM Popovich, Jr, PT, DPT, is a PhD student in biomechanics, Musculoskeletal Biomechanics Research Laboratory, Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, Calif.

GC Gracey, PT, MPT, CPed, is Physical Therapist, Contractor, Medical Section, Department of Rehabilitation Medicine, National Institutes of Health.

JV Danoff, PT, PhD, is Research Consultant, Physical Therapy Section, Department of Rehabilitation Medicine, National Institutes of Health, and Associate Professor, Department of Exercise Science, School of Public Health, George Washington University, Washington, DC.

All authors provided concept/idea/research design, writing, and data analysis. Mr Shrader, Dr Popovich, and Mr Gracey provided data collection. Mr Shrader provided project management. The authors thank Dr Lynn H Gerber for her mentorship and support of this project.

This study was approved by the institutional review board of the National Cancer Institnte at the National Institutes of Health.

The opinions and information contained in this article are those of the authors and do not necessarily reflect those of the Department of Health and Human Services, National Institutes of Health, or the US Public Health Service.

This article was received September 22, 2004, and was accepted January 7, 2005.

Table 1.
Navicular Drop Values for Subjects Without Foot and Ankle Disorders in
Investigations With Similar Methods Reporting Intraclass Correlation
Coefficients (ICCs) of [greater than or equal to] 80 for Intrarater
Reliability

                        No. of   Navicular   Intrarater    Interrater
                        Feet     Drop        Reliability   Reliability
Authors/Year            Tested   (mm)        (ICC [2,1))   (ICC [2,1])

Allen and Glasoe,         36     8.1         .90
  (16) 2000
McPoil and Cornwall,      56     6.2         .94-.98 (a)
  (13) 1996
Sell et al, (14) 1994     60     6.0         .83               .73
Mueller et al,            58     7.3         .78-.83
  (29) 1993

(a) Investigated reliability o^ navicular height measurements from
subtalar joint neutral and relaxed standing foot postures but did not
include navicular drop calculation reliability.

Table 2.

Foot Function Index (FFI), American College of Rheumatology (ACR)
Functional Classification,33 and Number of Subjects With Foot Pain
(N=10)

                                                [bar.X]    SD
FFI total (0-100 scale; lower scores=lower        24.0    14.5
  pain and impairment)
FFI pain subscale                                 29.3    21.6
FFI disability subscale                           32.3    20.9
FFI activity limitation subscale                  10.3    10.7
No. of subjects with ACR class I                  10
No. of subjects experiencing foot pain during      7
  the past week

Table 3.

American College of Rheumatology (ACR) Classification of Global
Functional Status in Rheumatoid Arthritis (a)

ACR

Classification   Description

Class I          Able to perform usual activities of daily living
                   (self-care, vocational, and avocational)
Class II         Able to perform usual self-care and
                   vocational activities, but limited in
                   avocational activities

Class III        Able to perform usual self-care activities, but
                   limited in vocational and avocational
                   activities

Class IV         Limited in ability to perform usual self-care,
                   vocational, and avocational activities

"Reprinted with permission of Wiley-Liss Inc, a subsidiary of John
Wiley & Sons Inc, from: Hochberg MC, Chang RW, Dwosh I, et al. The
American College of Rheumatology 1991 revised criteria for the
classification of global functional status in rheumatoid arthritis.
Arthritis Rheum. 1992;35:498-502. Copyright 1992.

Table 4.

Means and Standard Deviations for All Navicular Height (NH)
Measurements and Navicular Drop (ND) Calculations (a)

                               Session 1       Session 2
                              X        SD    X        SD

(NH 1) STJN                   44.71   5.60   44.05   5.45
(NH2) RSFP                    36.34   7.68   35.76   7.44
(NH3) RSFP with repalpation   34.76   7.92   34.48   7.55
(NH4) SLS                     35.17   7.25   34.75   6.81
ND1=(NH1-NH2)                  8.36   5.29    8.29   5.24
ND2=(NH1-NH3)                  9.95   5.44    9.57   5.37
ND3=(NH1-NH4)                  9.42   5.04    9.30   5.32

(a) STJN=subtalar joint neutral, RSFP=relaxed standing foot posture,
SLS=single-limb stance, ND 1=navicular drop as reported in the
literature, ND2=navicular drop with account of skin marking error,
ND3=navicular drop as difference between bilateral standing in STJN
and SLS in RSFP.

Table 5.

Interrater Reliability: Intraclass Correlation Coefficients (ICCs) and
Standard Errors of Measurement (SEMs) (a)

                Session 1
                ICC (2,1)   SEM    ICC (2,k)   SEM

(NH 1) STJN     .77         3.33   .91         1.61
(NH2) RSFP      .86         3.23   .95         1.68
(NH3) RSFP      .9          3.06   .96         1.54
  with
  repalpation
(NH4) SLS       .84         3.20   .94         1.63
ND1=(NH1-NH2)   .71         2.62   .88         1.81
ND2=(NH1-NH3)   .79         2.93   .91         1.61
ND3=(NH1-NH4)   .67         3.20   .85         1.96

                Session 2
                ICC (2,1)   SEM    ICC (2,k)   SEM

(NH 1) STJN     .85         2.46   .94         1.29
(NH2) RSFP      .92         2.31   .97         1.20
(NH3) RSFP      .92         2.17   .97         1.29
  with
  repalpation
(NH4) SLS       .88         2.25   .96         1.36
ND1=(NH1-NH2)   .83         2.14   .94         1.24
ND2=(NH1-NH3)   .90         1.98   .96         1.07
ND3=(NH1-NH4)   .85         2.37   .95         1.19

(a) STJN=subtalar joint neutral, RSFP=relaxed standing foot posture,
SLS=single-limb stance, ND 1=navicular drop as reported in the
literature, ND2=navicular drop with account of skin marking error,
ND3=navicular drop as difference between bilateral standing in STJN
and SLS in RSFP.

Table 6.

Intrarater Reliability: Intraclass Correlation Coefficients (ICCs) and
Standard Errors of Measurement (SEMs) (a)

                Examiner 1
                ICC (2,1)    SEM    ICC (2,k)   SEM

NH 1) STJN      .84          1.61   .94         1.11
(NH2) RSFP      .92          2.09   .98         0.94
(NH3) RSFP      .94          1.85   .98         0.97
  with
  repalpation
(NH4) SLS       .95          1.57   .97         0.89
ND2=(NH1-NH3)   .90          1.89   .98         0.72
ND3=(NH1-NH4)   .87          2.04   .98         0.71

                Examiner 2
                ICC (2,1)    SEM     ICC (2,k)   SEM

NH 1) STJN      .87          2.27    .95         1.34
(NH2) RSFP      .94          1.71    .97         1.20
(NH3) RSFP      .93          2.05    .97         1.24
  with
  repalpation
(NH4) SLS       .92          1.92    .98         0.91
ND1=(NH1-NH2)   .88          1.80    .95         1.14
ND2=(NH1-NH3)   .75          2.46    .93         1.23
ND3=(NH1-NH4)   .73          2.47    .94         1.13

                Examiner 3
                ICC (2,1)    SEM    ICC (2,k)   SEM

NH 1) STJN      .79          2.75   .95         1.16
(NH2) RSFP      .90          2.37   .97         1.37
(NH3) RSFP      .92          2.22   .98         1.07
  with
  repalpation
(NH4) SLS       .85          2.66   .95         1.56
ND1=(NH1-NH2)   .75          2.25   .96         0.82
ND2=(NH1-NH3)   .83          2.24   .97         0.90
ND3=(NH1-NH4)   .80          2.27   .90         1.45

(a) STJN=subtalar joint neutral, RSFP=relaxed standing foot posture,
SLS=single-limb stance, NDI=navicular drop as reported in the
literature, ND2=navicular drop with account of skin marking error,
ND3=navicular drop as difference between bilateral standing  in STJN
and SLS in RSFP.