Reliability of gait measurements in people with osteoarthritis of the knee.Key Words: Gait analysis gait analysis Rehab medicine Evaluation of the gait of Pts with a neurologic or orthopedic condition affecting the motor control system–eg, brain injury, spinal cord injury, cerebral palsy, stroke, multiple sclerosis, musculoskeletal actuator systems, post , Knee, Osteoarthritis osteoarthritis or osteoarthrosis or degenerative joint disease Most common joint disorder, afflicting over 80% of those who reach age 70. It does not involve excessive inflammation and may have no symptoms, especially at first. , Reliability. Gait gait (gat) the manner or style of walking. antalgic gait a limp adopted so as to avoid pain on weight-bearing structures, characterized by a very short stance phase. is a highly complex activity in which many variables can be observed and measured. There is almost no limit to the complexity of data that can be collected, but complex data usually require costly equipment and time-consuming data acquisition and analysis[1,2] Hirokawa and Matsumara[3] suggest that because gait is highly individual, the most important data derived are the discrete variables Discrete variable Variable like 1, 2, 3. Bond ratings are examples of discrete classifications. representing its final output: walking speed, cadence cadence, in music, the ending of a phrase or composition. In singing the voice may be raised or lowered, or the singer may execute elaborate variations within the key. , and stride length stride length Biomechanics The distance between 2 successive placements of the same foot, consisting of 2 step lengths; SL measured between successive positions of the left foot is always the same as that measured by the right foot, unless the subject is walking in a curve . These variables are particularly relevant to patients with osteoarthritis involving the lower limbs, because decreases in walking speed and stride length have been demonstrated in many studies involving such individuals.[4-6] To be able to test large clinical populations, it is essential to keep the assessment procedure as concise and easy to perform as possible for both the assessor and the patient. Although we have validated self-report measures of pain and physical function for persons with osteoarthritis of the knee,[7-10] additional measures are needed as outcome measures in intervention studies intervention studies, n.pl the epidemiologic investigations designed to test a hypothesized cause and effect relation by modifying the supposed causal factor(s) in the study population. . Osteoarthritis of the knee is characterized in the early stages by a level of joint pain with no reports of functional limitations. Alternatively, some individuals may report little pain because they have adjusted the means of performance of their daily activities.[7] although the currently most cited self-report physical function assessments used for patients with osteoarthritis of the knee--the Western Ontario and McMaster Universities McMaster University, at Hamilton, Ont., Canada; nondenominational; founded 1887. It has faculties of humanities, science, social sciences, business, engineering, and health sciences, as well as a school of graduate studies and a divinity college. Osteoarthritis Index[8] (WOMAC WOMAC Western Ontario McMaster University Osteoarthritis Index Rheumatology An arthritic pain scoring system ranging from 0–no pain/disability to 100–most severe pain/disability ) and the Stanford Health Assessment Questionnaire[9,10] (HAQ HAQ Health Assessment Questionnaire HAQ Harvard Asia Quarterly )--have been validated for this population, they have not been validated for individuals in the early stages of the disease. Changes in gait variables may occur in many people before any functional loss occurs, as scored by the WOMAC or the HAQ or even before pain is recorded by visual analog scales (VASs). Gibbs et al[11] have suggested, after a cross-sectional study cross-sectional study n. See synchronic study. cross-sectional study, n the scientific method for the analysis of data gathered from two or more samples at one point in time. of 532 subjects, that decreased gait speed is a precursor precursor /pre·cur·sor/ (pre´kur-ser) something that precedes. In biological processes, a substance from which another, usually more active or mature, substance is formed. In clinical medicine, a sign or symptom that heralds another. of functional limitations. Gait speed, therefore, seems to be an outcome measure that is potentially useful in studies assessing the efficacy of therapeutic intervention in this population. An inexpensive method of collecting gait data reliably for these patients using footprint recordings and a stopwatch was reported by Falconer Falconer prison where former professor Farragut, who had killed his brother, witnesses the torments and chaos of the penal system. [Am. Lit.: Cheever Falconer in Weiss, 151] See : Imprisonment and Hayes.[12] This method, however, can be impractical im·prac·ti·cal adj. 1. Unwise to implement or maintain in practice: Refloating the sunken ship proved impractical because of the great expense. 2. for repeated testing of large clinical populations or several gait trials on one subject because the variables need to be measured manually. An electronic footswitch system, due to its ease of use for both the assessor and the patient, can be used to test large numbers of patients or several trials while providing measurements of mean walking speed, cadence, and stride length. In our cost-sensitive health system, medical and paramedical par·a·med·i·cal adj. 1. Of, relating to, or being a person trained to give emergency medical treatment or assist medical professionals. 2. interventions need to demonstrate efficacy. If quantitative gait analysis is to be used as an outcome measure for therapeutic intervention or as a tool to assess an individual's deviation from normal or to document disease progression (eg, progression of arthritis), the level of reliability for each collected variable needs to be established. Four main types of reliability have been identified: instrument, intrarater, interrater, and intrasubject.[13] The nature of a footswitch system, with its automated data collection and analysis used together with a standardized testing A standardized test is a test administered and scored in a standard manner. The tests are designed in such a way that the "questions, conditions for administering, scoring procedures, and interpretations are consistent" [1] protocol, could minimize factors affecting the first three types of reliability. Of particular interest in persons with arthritis is intrasubject reliability or changes that occur because of variability in subject performance without intervention. In a study using a footswitch system, Heussler[14] demonstrated that results from subjects without impairments are repeatable over 1 week for walking speed (F=2.13; df=1,18; P=.15) and stride length (F=1.88; df=1,18; P=.17). Marks,[15] who examined only gait speed in a small sample of subjects with osteoarthritis of the knee, claimed reliability across a period of 1 week with intraclass correlation In statistics, the intraclass correlation (or the intraclass correlation coefficient[1]) is a measure of correlation, consistency or conformity for a data set when it has multiple groups. coefficients (ICC ICC See: International Chamber of Commerce [1,1]) of.80 to .88. The study, however, had limited external validity External validity is a form of experimental validity.[1] An experiment is said to possess external validity if the experiment’s results hold across different experimental settings, procedures and participants. because the majority of subjects were stabilized with nonsteroidal anti-inflammatory drugs Nonsteroidal Anti-Inflammatory Drugs Definition Nonsteroidal anti-inflammatory drugs are medicines that relieve pain, swelling, stiffness, and inflammation. and were only moderately impaired. The level of intrasubject reliability, therefore, has not been established for the quantitative gait variables of walking speed, cadence, or stride length in patients with osteearthritis. This study was undertaken to determine the minimum number of trials needed per assessment session to establish stable measurements and optimal intersession in·ter·ses·sion n. The time between two academic sessions or semesters. in  ter·ses reliability.Method Subjects All patients with a diagnosis of osteoarthritis of the knee on the physical therapy waiting list of a large public hospital were invited to participate. Measurements were taken from 41 patients who fulfilled the clinical and radiological radiological pertaining to radiology. radiological diagnosis see radiological diagnosis. mobile radiological apparatus x-ray machines that can be moved but are not portable because of their weight. criteria of the American College American College is the name of:
rheu·ma·tol·o·gy n. for the diagnosis of osteoarthritis of the knee.[16] The patient characteristics are shown in Table 1. Patients were excluded if they (1) had any large effusions or joint arthroplasty in the lower limbs, (2) had neurological disorders This is a list of major and frequently observed neurological disorders (e.g. Alzheimer's disease), symptoms (e.g.back pain), signs (e.g. aphasia) and syndromes (e.g. Aicardi syndrome). affecting gait, (3) had received intra-articular injections to the lower-limb joints within the previous 2 months, or (4) were not stabilized with previously prescribed pre·scribe v. pre·scribed, pre·scrib·ing, pre·scribes v.tr. 1. To set down as a rule or guide; enjoin. See Synonyms at dictate. 2. To order the use of (a medicine or other treatment). medications for at least 3 weeks prior to the initial assessment.
Table 1.
Descriptive Characteristics of Subjects
Men (n=6)
[bar] X SD Range
Age (y) 68.0 7.5 54-75
Height (cm) 180.8 8.5 169-190
Weight (kg) 95.7 10.7 84-115
Symptom duration (y) 3.5 3.5 1-10
Women (n=35)
[bar] X SD Range
Age (y) 68.0 7.4 50-85
Height (cm) 156.6 5.4 144-168
Weight (kg) 73.6 13.1 53-110
Symptom duration (y) 4.5 3.7 1-12
The subjects had a mean symptom duration of approximately 4 years (Tab. 1), with only 6 subjects having unilateral unilateral /uni·lat·er·al/ (-lat´er-al) affecting only one side. u·ni·lat·er·al adj. On, having, or confined to only one side. disease. None of the subjects were using a walking aid for indoor ambulation am·bu·late intr.v. am·bu·lat·ed, am·bu·lat·ing, am·bu·lates To walk from place to place; move about. [Latin ambul or were receiving any physical therapy during the assessment period. All eligible subjects agreed to participate and signed an informed consent form. Instrumentation Gait was assessed using an 8-m electronic footswitch walkway walkway Rehabilitation medicine An instrument used to measure the timing of foot contact and or position of the foot on the ground . The gait variables that were measured were mean walking speed (in centimeters per second), cadence (in steps per minute), and mean stride length (in centimeters). The footswitch system, which has a sampling rate of 100 Hz, is based on a footswitch system designed by Larsson et al[17] and is similar to the footswitch system described by Whittle.[2] The system consists of a level walkway, footswitches, trailing leads supported by an overhead pulley pulley, simple machine consisting of a wheel over which a rope, belt, chain, or cable runs. A grooved pulley wheel like that used for ropes is called a sheave. system running on a low-friction aluminum curtain track, active microoptical sensors positioned at 3 and 5 m along the walkway, a signal-conditioning box, and a microcomputer microcomputer Small digital computers whose CPU is contained on a single integrated semiconductor chip. As large-scale and then very large-scale integration (VLSI) have progressively increased the number of transistors that can be placed on one chip, the processing capacity (Fig. 1). To eliminate the distorting effect of naturally occurring gait acceleration and deceleration deceleration /de·cel·er·a·tion/ (de-sel?er-a´shun) decrease in rate or speed. early deceleration ,[18] the software analyzes the gait from the data recorded after the first microoptical beam has been broken. The two microoptical sensors, positioned at shoulder level exactly 200 cm apart, are used to measure the subject s walking speed. The maximum response time of the sensors is 1.5 milliseconds. Cadence and stride length are determined from the current generated when the 57-mm-long brass clip footswitches attached to the subject s shoes contact the walkway (Fig. 2). Cadence and stride length are analyzed an·a·lyze tr.v. an·a·lyzed, an·a·lyz·ing, an·a·lyz·es 1. To examine methodically by separating into parts and studying their interrelations. 2. Chemistry To make a chemical analysis of. 3. from the two strides occurring (from heel-switch) after the first microoptical beam has been broken. The brass strips are attached to the shoes at the toe, approximately at the level of the base of the fifth metatarsal metatarsal /meta·tar·sal/ (met?ah-tahr´sal) 1. pertaining to the metatarsus. 2. a bone of the metatarsus. met·a·tar·sal adj. Of or relating to the metatarsus. , and centrally at the back of the heel. The height of the strips is approximately 0.01 mm. When combined with the double-sided adhesive, the strips are less than 0.04 mm high without compression, which ensures minimal disturbance to the subject's gait. [Figure 1 and 2 ILLUSTRATION OMITTED] The analog signals An analog or analogue signal is any time continuous signal where some time varying feature of the signal is a representation of some other time varying quantity. It differs from a digital signal in that small fluctuations in the signal are meaningful. from the footswitches and the microoptical sensors are smoothed and filtered by a threshold detector with a positive feedback loop before being fed into the microcomputer for conversion to digital form. The positive feedback loop creates a threshold window and eliminates multiple triggering, which could occur during the transitional periods when the signal is changing. Software programs written specifically for the system then extract the relevant gait variables. Protocol Three measurements were made at 1-week intervals at approximately the same time of day (+/-1 hour)[19] The testing sequence was nonrandom. Subjects initially completed self-report assessments of pain during level walking (100-mm VASs) and physical function (Stanford Health Assessment Questionnaire).[9,10] The footswitches were fixed to the subjects' shoes, with the trailing leads attached to the subjects' clothing. One initial test walk only was undertaken at each session at an unspecified Adj. 1. unspecified - not stated explicitly or in detail; "threatened unspecified reprisals" specified - clearly and explicitly stated; "meals are at specified times" speed to control for possible hindrances to ambulation. Subjects were initially asked to "walk at a pace that you consider to be normal" at each of five trials and then to "walk at a pace that you consider to be fast" at each of the next five trials. Subjects had a 45- to 60-second standing pause between trials. No other cuing or verbal encouragement was given. The subjects wore their own same low-heeled shoes. If pain-relieving medications were being used, the time interval between taking the medication and each measurement session was kept constant. Data Analysis Gait speed, cadence, and stride length were analyzed at the normal and fast self-selected walking speeds to evaluate (1) the level of intrasession reliability, (2) the intersession reliability to establish the optimum number of gait trials, up to a maximum of five trials, needed at each measurement session for each self-selected walking speed, and (3) the minimum change needed for the clinician clinician /cli·ni·cian/ (kli-nish´in) an expert clinical physician and teacher. cli·ni·cian n. to account for measurement error. The significance level was not lowered for multiple comparisons, as this was a reliability study and therefore the priority was to avoid concluding that there was no change between measurements when a change did exist (Type II error). A univariate repeated-measures analysis of variance (ANOVA anova see analysis of variance. ANOVA Analysis of variance, see there ) was performed on the intrasession gait data to assess the statistical significance of any changes. An ANOVA was also performed on the intersession walking pain and physical function scores to assess the level of symptom stability over the weeks. Adjustments were made to the ANOVA results, using the Greenhouse-Geisser Epsilon 1. (language) EPSILON - A macro language with high level features including strings and lists, developed by A.P. Ershov at Novosibirsk in 1967. EPSILON was used to implement ALGOL 68 on the M-220. , when Mauchly's sphericity test Mauchly's sphericity test is a statistical test used to validate repeated measures factor ANOVA. SPSS includes it automatically as part of its ANOVA with a repeated measures factor (RMF) output tables. The test was introduced by ENIAC co-inventor John Mauchly in 1940. assumptions were violated.[20] Only statistically significant results were examined (P [is less than] .05), as the significance level for the corrected test will always be higher than unadjusted results. Intraclass correlation coefficients (ICC[2,1]) were calculated to determine the relative reliability of the intrasession and intersession measurements.[21] The ICC(2,1) was chosen so that the results could be extrapolated to other judges. To determine the measurement error in metric terms, the standard error of measurement (SEM) was also calculated for intersession data.[13] The SEM gives an estimate of the minimum change needed in an individual to account for measurement error. To further aid the identification of the relative merits of the different gait variables in absolute terms (Alg.) such as are known, or which do not contain the unknown quantity. See also: Absolute , intersession percentages of agreement were also calculated. Data were analyzed with the Statistical Package for the Social Sciences (statistics, tool) Statistical Package for the Social Sciences - (SPSS) The flagship program of SPSS, Inc., written in the late 1960s. ["SPSS X User's Guide", SPSS, Inc. 1986]. (SPSS/PC+ version 4.0(*)) and ICC.EXE Exe (ĕks), river, c.55 mi (90 km) long, rising in the Exmoor, Somerset, SW England, and flowing S across the Cornwall peninsula, past Exeter to the English Channel at Exmouth. software.[dagger] Results The mean VAS vas (vas) pl. va´ sa [L.] vessel.va´sal vas aber´rans 1. a blind tubule sometimes connected with the epididymis; a vestigial mesonephric tubule. 2. walking pain scores (0 representing no pain and 100 representing maximum pain) and the disability index derived from the HAQ (ranging from 0 to 3, with higher scores representing greater disability) for the three measurement sessions are given in Table 2. The ANOVA demonstrated no changes between the three sessions for either score and established symptom and physical function stability in this sample over the assessment period. Table 2. Level-Walking Pain and Stanford Health Assessment Questionnaire[9,10](HAQ) Score
Session 1 Session 2 Session 3
X SD X SD X SD
Walking pain
(0-100 mm) 35.4 31.7 30.0 29.5 31.7 3.0
HAQ (0-3) 0.64 0.36 0.62 0.39 0.58 0.41
F P ICC(2, 1)(a)
Walking pain 1.68 .19 .62
(0-100 mm) 2.73 .07 .90
HAQ 10-3)
(a) =intraclass correlation coefficient. The descriptive statistics descriptive statistics see statistics. for the quantitative gait variables at each of the three measurement sessions are presented in Table 3. Although the electronic footswitch system used in our study provided analysis in centimeters per second, steps per minute, and centimeters for walking speed, cadence, and stride length, respectively, the data were converted to meters per second, steps per second, and meters to be more comparable to much of the published literature to date.
Table 3. Descriptive Statistics of Quantitative Gait Variables
for Trials 1 Through 5 Within Each Session
Session 1 Session 2
X SD X SD
Normal walking speed
(m [multiplied by] [s.sup.-1] 0.88 0.22 0.95 0.23
0.92 0.21 0.96 0.22
0.94 0.21 0.98 0.22
0.94 0.21 0.98 0.21
0.95 0.22 0.95 0.23
Normal cadence
(steps [multiplied by]
[s.sup.-1] 1.61 0.21 1.71 0.20
1.66 0.20 1.72 0.20
1.67 0.20 1.73 0.19
1.66 0.20 1.73 0.19
1.69 0.19 1.75 0.19
Normal stride length (m) 1.08 0.19 1.10 0.20
1.10 0.20 1.10 0.19
1.10 0.21 1.13 0.19
1.12 0.19 1.13 0.19
1.12 0.19 1.13 0.19
Fast walking speed
(m [multiplied by] [s.sup.-1] 1.18 0.28 1.17 0.24
1.19 0.25 1.18 0.23
1.20 0.25 1.19 0.23
1.19 0.24 1.21 0.24
1.21 0.24 1.21 0.23
Fast cadence
(steps (m [multiplied by]
[s.sup.-1] 1.91 0.21 1.94 0.19
1.92 0.18 1.93 0.19
1.92 0.19 1.93 0.20
1.91 0.18 1.94 0.19
1.93 0.17 1.96 0.19
Fast stride length (m) 1.22 0.22 1.21 0.21
1.23 0.21 1.22 0.20
1.24 0.20 1.23 0.19
1.24 0.22 1.24 0.19
1.25 0.21 1.24 0.19
Session 3
X SD
Normal walking speed
(m [multiplied by] [s.sup.-1] 0.98 0.20
0.99 0.20
0.99 0.20
1.00 0.20
1.01 0.20
Normal cadence
(steps [multiplied by] [s.sup.-1] 1.72 0.19
1.75 0.21
1.75 0.19
1.75 0.20
1.77 0.17
Normal stride length (m) 1.13 0.18
1.13 0.17
1.13 0.17
1.14 0.17
1.14 0.17
Fast walking speed
(m [multiplied by] [s.sup.-1] 1.21 0.22
1.20 0.22
1.22 0.20
1.22 0.22
1.23 0.22
Fast cadence
(steps (m [multiplied by]
[s.sup.-1] 1.98 0.21
1.96 0.19
1.97 0.19
1.99 0.21
2.00 0.21
Fast stride length (m) 1.22 0.18
1.22 0.19
1.23 0.18
1.22 0.19
1.23 0.17
Intrasession Reliability Data for three different assessment protocols were analyzed: (1) trials 1 through 5, (2) trials 2 through 5, and (3) trials 2 and 3. Table 4 shows that the mean intrasession ICC(2,1) values for walking speed, cadence, and stride length were consistently high (ICC=.90 - .98), with the lower boundaries of the 95% confidence intervals confidence interval, n a statistical device used to determine the range within which an acceptable datum would fall. Confidence intervals are usually expressed in percentages, typically 95% or 99%. ranging from .84 to .96. The ANOVA, however, revealed that there were intrasession differences at the normal self-selected walking speed for trials 1 through 5. When trial 1 is excluded from the analysis, cadence and stride length demonstrate no intrasession differences. In contrast, at the fast self-selected walking speed, there were no intrasession differences for any of the assessment protocols for walking speed, cadence, and stride length. Table 4. Results of Repeated-Measures Analysis of Variance With Intraclass Correlation Coefficients (ICC[2,1]) and 95% Confidence Intervals (Cls)
Trials F P ICC
Normal walking speed
(m [multiplied by]
[s.sup.-1] 1-5 16.81 <.001 .93
2-5 4.85 .003 .96
2, 3 5.07 .030 .96
Normal cadence
(steps [multiplied by]
[s.sup.-1] 1-5 10.38 <.001 .90
2-5 1.28 .285 .92
2, 3 0.97 .600 .92
Normal stride length (m) 1-5 4.31 .002 .93
2-5 1.62 .187 .94
2, 3 0.00 .969 .95
Fast walking speed
(m [multiplied by] 1-5 1.60 .176 .93
[s.sup.-1] 2-5 2.21 .091 .95
2, 3 3.00 .091 .98
Fast cadence
(steps (m [multiplied by] 1-5 0.83 .507 .90
[s.sup.-1] 2-5 1.00 .396 .91
2, 3 0.01 .906 .91
Fast stride length (m) 1-5 2.33 .045 .94
2-5 1.74 .162 .95
2, 3 0.71 .405 .94
95% Cl
Normal walking speed
(m [multiplied by]
[s.sup.-1] .88-.96
.89-.96
.93-.98
Normal cadence
(steps [multiplied by]
[s.sup.-1] .84-.94
.88-.95
.86-.96
Normal stride length (m) .90-.96
.90-.96
.91-.97
Fast walking speed
(m [multiplied by] .90-.96
[s.sup.-1] .93-.97
.96-.99
Fast cadence
(steps (m [multiplied by] .85-.94
[s.sup.-1] .86-.95
.84-.95
Fast stride length (m) .91-.97
.92-.97
.89-.97
Intersession Reliability Using the same assessment protocols (trials 1-5, trials 2-5, and trials 2 and 3), data were analyzed to establish the presence and characteristics of intersession reliability. These results are given in Table 5. Compared with the intrasession results shown in Table 4, ICC(2,1) values for the first intersession analysis (sessions 1 and 2) remained stable for stride length at both self-selected walking speeds but decreased for walking speed and cadence, most markedly at the normal self-selected walking speed. This trend is particularly evident when comparing the lower boundaries of the 95% confidence intervals. These findings reflect, in part, the additive additive In foods, any of various chemical substances added to produce desirable effects. Additives include such substances as artificial or natural colourings and flavourings; stabilizers, emulsifiers, and thickeners; preservatives and humectants (moisture-retainers); and value of the mean scores from session 1 to session 2 for these variables. There were mean increases from session 1 to session 2 of 0.04 m [multiplied by] [s.sup.-1] for normal walking speed and 0.07 steps [multiplied by] [s.sup.-1] for normal cadence. In contrast, there was a mean increase of only 0.01 m for normal stride length, a mean decrease of -0.01 m [multiplied by] [s.sup.-1] for fast walking speed, a mean increase of 0.02 steps [multiplied by] [s.sup.-1] for fast cadence, and a mean decrease of 0.01 m for fast stride length from session 1 to session 2. The second intersession analysis (sessions 2 and 3) showed a consistent increase of the ICC(2,1) values.
Table 5.
Mean Intersession Intraclass Correlation Coefficients
(ICC[2,1]) With 95% Confidence Intervals (Cls)
Trials Per Sessions 1 and 2
Session ICC 95% Cl
Normal walking speed
(m [multiplied by]
[s.sup.-1] 1-5 .88 .79-.94
2-5 .89 80-.94
2, 3 .88 78-.93
Normal cadence
(steps [multiplied by]
[s.sup.-1] 1-5 .79 .52-.90
2-5 .81 .58-.91
2, 3 .78 .58-.88
Normal stride length (m) 1-5 .94 .89-.97
2-5 .94 .89-.97
2, 3 .93 .87-.96
Fast walking speed 1-5 .91 .84-.95
(m [multiplied by] 2-5 .92 .85-.96
[s.sup.-1] 2, 3 .91 .84-.95
Fast cadence
(steps (m [multiplied by]
[s.sup.-1] 1-5 .85 .74-.92
2-5 .84 .71 -.91
2, 3 .85 .74-.92
Fast stride length (m) 1-5 .93 .87-.97
2-5 .94 .90-.97
2, 3 .93 .87-.96
Sessions 2 and 3
ICC 95% Cl
Normal walking speed
(m [multiplied by]
[s.sup.-1] .94 .88-.97
.94 .88-.97
.94 .89-.97
Normal cadence
(steps [multiplied by]
[s.sup.-1] .85 .73-.92
.84 .72-.91
.81 .67-.90
Normal stride length (m) .96 .92-.98
.95 .91-.98
.94 .88-.97
Fast walking speed .95 .91-.98
(m [multiplied by] .95 .91-.97
[s.sup.-1] .95 .90-.97
Fast cadence
(steps (m [multiplied by]
[s.sup.-1] .90 .80-.95
.90 .80-.95
.89 .79-.94
Fast stride length (m) .98 .95-.99
.97 .94-.98
.96 .92-.98
The ICC(2,1) is an established measure of relative reliability, but it is very sensitive to the range of scores or intersubject variability. To supplement the ICCs, a measure of actual agreement among measurement sessions was determined. The scores shown in Table 6 are the cumulative percentages of agreement for 0.02, 0.04, and 0.06 units of the variables. The decision to choose these units was based on the intrasession SEM values, which ranged from 0.03 to 0.06 units. At the normal walking speed for trials 1 through 5, for example, there was only 41% agreement within 0.06 m [multiplied by] [s.sup.-1] between sessions 1 and 2. The percentage of agreement for this variable and protocol increased markedly to 72% between session 2 and 3. The results in Table 6 highlight again the consistent greater reliability demonstrated at the second intersession analysis (sessions 2 and 3) compared with the first intersession analysis (sessions 1 and 2). This trend for increased reliability at the second intersession analysis was particularly marked for normal self-selected walking speed and cadence. Table 6. Intersession Cumulative Percentages of Agreement for 0.02, 0.04, and 0.06 Units of the Variables
Trials Per Sessions 1 and 2
Session 0.02 0.04 0.06
Normal walking speed 1-5 14 24 41
(m [multiplied by] 2-5 10 28 46
[s.sup.-1]) 2, 3 18 28 44
Normal cadence 1-5 12 20 27
(steps [multiplied by] 2-5 12 22 41
[s.sup.-1]) 2, 3 17 24 32
Normal stride length 1-5 37 58 69
(m) 2-5 39 58 74
2, 3 32 39 68
Fast walking speed 1-5 17 32 51
(m [multiplied by] 2-5 29 46 54
[s.sup.-1]) 2, 3 29 46 54
Fast cadence 1-5 12 27 41
(steps [multiplied by] 2-5 15 29 37
[s.sup.-1]) 2, 3 24 39 41
Fast stride length 1-5 41 59 77
(m) 2-5 41 62 79
2, 3 36 62 73
Sessions 2 and 3
0.02 0.04 0.06
Normal walking speed 33 51 72
(m [multiplied by] 33 54 64
[s.sup.-1]) 26 46 72
Normal cadence 37 56 61
(steps [multiplied by 44 51 59
[s.sup.-1]) 27 46 59
Normal stride length 42 71 79
(m) 32 63 84
26 55 68
Fast walking speed 30 55 75
(m [multiplied by] 30 50 63
[s.sup.-1]) 23 48 75
Fast cadence 24 41 59
(steps [multiplied by 27 32 58
[s.sup.-1]) 32 46 49
Fast stride length 60 78 93
(m) 50 73 90
43 60 80
Confidence Intervals The 90% confidence intervals presented in Table 7 demonstrate that at the normal self-selected walking speed in a simple preintervention-postintervention assessment and using gait measurements, increases of approximately 0.12 m [multiplied by] [s.sup.-1] for walking speed, 0.13 steps. [s.sup.-1] for cadence, and 0.08 m for stride length (depending on the number of trials used) would have to be found to demonstrate change above what may be attributable to measurement error. At the fast self-selected walking speed, increases of approximately 0.11 m [multiplied by] [s.sup.-1] for walking speed, 0.11 steps [multiplied by] [s.sup.-1] for cadence, and 0.08 m for stride length would be needed. These confidence intervals decreased markedly and consistently at the second intersession analysis.
Table 7. Intersession Standard Error of Measurement 90%
Confidence Intervals for the Clinician
Session Sessions
Trials 1 and 2 2 and 3
Normal walking speed 1-5 [+ or -] 0.12 [+ or -] 0.08
2-5 [+ or -] 0.12 [+ or -] 0.08
2, 3 [+ or -] 0.12 [+ or -] 0.08
Normal cadence 1-5 [+ or -] 0.12 [+ or -] 0.11
2-5 [+ or -] 0.12 [+ or -] 0.11
2, 3 [+ or -] 0.14 [+ or -] 0.12
Normal stride length 1-5 [+ or -] 0.07 [+ or -] 0.06
2-5 [+ or -] 0.08 [+ or -] 0.06
2, 3 [+ or -] 0.08 [+ or -] 0.07
Fast walking speed 1-5 [+ or -] 0.12 [+ or -] 0.08
2-5 [+ or -] 0.11 [+ or -] 0.08
2, 3 [+ or -] 0.12 [+ or -] 0.08
Fast cadence 1-5 [+ or -] 0.11 [+ or -] 0.09
2-5 [+ or -] 0.11 [+ or -] 0.09
2, 3 [+ or -] 0.11 [+ or -] 0.09
Fast stride length 1-5 [+ or -] 0.08 [+ or -] 0.05
2-5 [+ or -] 0.07 [+ or -] 0.05
2, 3 [+ or -] 0.08 [+ or -] 0.05
Discussion Walking pain and physical function scores were collected at each measurement session to ascertain whether these scores were stable for the subjects studied. Without stability in these self-report scores, reliability of the objectively measured gait variables would have led to the validity or the responsiveness to change of quantitative gait analysis being questioned. It has been established at The St George Hospital Saint George may refer to:
adj. Exhibiting or producing no symptoms. Asymptomatic Persons who carry a disease and are usually capable of transmitting the disease but, who do not exhibit symptoms of the disease are said to be controls.[22] It should be noted, however, that none of the subjects in our study needed walking aids for indoor ambulation, had any large joint effusions or joint arthroplasty in the lower limbs, and were able to increase their gait speed by approximately 28% when asked to walk at a pace that they considered to be fast. This finding would indicate that, for most subjects in this study, walking at a pace that they considered to be normal was within their comfortable range. The several main trends discussed in this section, therefore, should not be extrapolated universally to persons with arthritis. It should also be noted that these reliability results are dependent on the standardized standardized pertaining to data that have been submitted to standardization procedures. standardized morbidity rate see morbidity rate. standardized mortality rate see mortality rate. protocol outlined in the "Method" section. The first major trend found in this study was the superior intrasession and intersession reliability of quantitative gait analysis at the fast self-selected walking speed when compared with the normal self-selected walking speed. The literature has documented the greater variability that is present with submaximal effort during muscle testing[23] and with slower functional movements.[24] This greater variability with submaximal effort has been attributed to the more complex recruitment coding and feedback requirements. In contrast, maximal max·i·mal adj. 1. Of, relating to, or consisting of a maximum. 2. Being the greatest or highest possible. effort or fast movements involve a simple all-out effort. This may be the reason for the greater intrasubject reliability at the fast self-selected walking speed in this sample and may explain why the fast self-selected walking speed provides the more stable measurement. The greater reliability at the fast walking speed, however, may have been influenced by the testing sequence. The five trials at the fast walking speed always followed the five trials at the normal walking speed. This sequence allows for a greater adaptability by the subject to the equipment. Conversely con·verse 1 intr.v. con·versed, con·vers·ing, con·vers·es 1. To engage in a spoken exchange of thoughts, ideas, or feelings; talk. See Synonyms at speak. 2. , the latter trials would be more influenced by fatigue or boredom Boredom See also Futility. Aldegonde, Lord St. bored nobleman, empty of pursuits. [Br. Lit.: Lothair] Baudelaire, Charles (1821–1867) French poet whose dissipated lifestyle led to inner despair. [Fr. Lit. , and the fast walking speed may demonstrate even greater reliability when placed first in the testing sequence. Future studies should examine the effect of varying the testing sequence. The optimum number of gait trials at each measurement session, within the analyzed five trials, was judged by analyzing the combined intersession reliabilities of walking speed, cadence, and stride length within a self-selected walking speed. Clearly, it would seem to be advisable ad·vis·a·ble adj. Worthy of being recommended or suggested; prudent. ad·vis a·bil to allow a familiarization fa·mil·iar·ize tr.v. fa·mil·iar·ized, fa·mil·iar·iz·ing, fa·mil·iar·iz·es 1. To make known, recognized, or familiar. 2. To make acquainted with. trial at each self-selected walking speed. In a simple preintervention-postintervention situation, the first intersession analysis (sessions 1 and 2) percentages of agreement shown in Table 6 suggest that no gain appears to be made by increasing the number of trials from two or three to two to five at either self-selected walking speed. A second major trend was the marked increase in reliability at the second intersession analysis (sessions 2 and 3) for both self-selected walking speeds (Tabs. 5-7). This increased reliability was particularly evident at the normal walking speed and cadence, indicating the decreasing influence of the learning or adaptability effect at this walking speed. These results would suggest that it is advisable to monitor the progress of a subject or of a group of subjects using a series of measurements because the decreasing measurement error and increasing reliability mean that smaller changes are needed to demonstrate real change. Our measurements were taken only 1 week apart, and the marked learning or adaptability effect noted at the normal walking speed might have been reduced if the time interval had been longer. For this reason, we suggest that the confidence intervals presented are conservative. The third major trend was that, within each self-selected walking speed, stride length was consistently the more stable intersession variable (Tabs. 5, 6). Females are more likely to increase their walking speed by increasing their cadence, whereas males are more likely to increase their walking speed by increasing their stride length.[18] The majority of our subjects were female (Tab. 1). This may be the reason for the greater intersession stability of stride length. Hirokawa,[18] however, also found stride length to be the most stable gait variable in a younger, asymptomatic male sample. Many outcome measures, such as pain and function, rely on self-reports. Quantitative gait analysis using an electronic footswitch system has the advantage of being an outcome measure that is not dependent on patient report. Apart from this important quality, the assessment is quick and easy to administer, and it is safe and nonthreatening even to persons with severe gait disabilities and those with cardiovascular compromise. In addition, only a few trials were needed to achieve excellent reliability for these measures (mean of two trials after a familiarization trial). When testing large groups of subjects or when dealing with subjects with gait disabilities, the added discomfort and time needed to collect data for multiple gait trials per measurement session or over multiple measurement sessions may not be sufficiently compensated for by increased measurement accuracy. If this is the case, then the clinician or researcher should be aware that a larger measurement error will affect the data collected. When time is crucial, then clearly analysis at the fast self-selected walking speed is superior to analysis at the normal self-selected walking speed. The results indicate, however, that one initial familiarization trial at each specified walking speed is essential prior to actual data collection. Conclusion This study demonstrated that quantitative gait analysis and a composite self-reported physical function score provided reliable baseline measurements for persons with osteoarthritis of the knee. For quantitative gait analysis, the study demonstrated that gait at the fast self-selected walking speed provided more reliable measurements than did gait at the normal self-selected walking speed, that reliability increased markedly from the first intersession analysis to the second intersession analysis, and that stride length was the most reliable quantitative gait variable within each self-selected walking speed. Although gait at the fast walking speed initially provided more reliable measurements when using quantitative gait analysis, it may not prove to be more responsive to change than gait at a normal walking speed. Acknowledgment acknowledgment, in law, formal declaration or admission by a person who executed an instrument (e.g., a will or a deed) that the instrument is his. The acknowledgment is made before a court, a notary public, or any other authorized person. We acknowledge the statistical advice given by Dr Roger Adams Roger Adams (January 2, 1889 – July 6, 1971) was an American organic chemist. He is best-known for the eponymous Adams' catalyst, but also greatly influenced graduate education in America, taught over 250 Ph.D. students and postgraduate students, and served the U.S. . (*) SPSS A statistical package from SPSS, Inc., Chicago (www.spss.com) that runs on PCs, most mainframes and minis and is used extensively in marketing research. It provides over 50 statistical processes, including regression analysis, correlation and analysis of variance. Inc, 444 N Michigan Ave, Chicago, IL 60611. ([dagger]) Public domain software available from Dr Roger Adams, Faculty of Health Sciences, The University of Sydney The University of Sydney, established in Sydney in 1850, is the oldest university in Australia. It is a member of Australia's "Group of Eight" Australian universities that are highly ranked in terms of their research performance. , PO Box 170, Lidcombe, New South Wales Lidcombe is a suburb in western Sydney, in the state of New South Wales Australia. Lidcombe is located 17 kilometres west of the Sydney central business district, in the local government area of Auburn Council. Lidcombe is colloquially known as ‘Liddy’. , Australia 2141. References [1] Perry J. Gait Analysis: Normal and Pathological 1. pathological - [scientific computation] Used of a data set that is grossly atypical of normal expected input, especially one that exposes a weakness or bug in whatever algorithm one is using. Function. Thorofare, NJ: Slack Inc; 1992. [2] Whittle M. Gait Analysis: An Introduction. Oxford, England: Butterworth & Co (Publishers) Ltd; 1991. [3] Hirokawa S, Matsumara K. Gait analysis using a measuring walkway for temporal and distance factors. Med Biol Eng Comput. 1987;25:577582. [4] Thin O, Pailhous Lafforgue P, Serratrice G. Quantitative analysis Quantitative Analysis A security analysis that uses financial information derived from company annual reports and income statements to evaluate an investment decision. Notes: of walking in patients with knee osteoarthritis: a method of assessing the effectiveness of non-steroidal anti-inflammatory treatment. Ann Rheum rheum (rldbomacm) any watery or catarrhal discharge. rheum n. A watery or thin mucous discharge from the eyes or nose. rheum any watery or catarrhal discharge. Dis. 1990;49:990-993. [5] Ivarsson I, Larsson L-E. Gait analysis in patients with gonarthrosis treated by high tibial osteotomy high tibial osteotomy Orthopedic surgery A procedure used for osteoarthritis in which a wedge of bone is excised from the tibial plate at the point of greatest contact with the femur; HTOs redistribute weight, and may ↓ cartilaginous wear . Clin Orthop. 1989;239:185-190. [6] Kroll M, Otis J, Sculco T, et al. The relationship of stride characteristics to pain before and after total knee arthroplasty. Clin Orthop. 1989;239:191-195 [7] Guccione AA, Felson DT, Anderson JJ. Defining arthritis and measuring functional status in elders: methodological issues in the study of disease and physical disability. Am J Public Health. 1990;80:945-949. [8] Bellamy N, Buchanan WW, Goldsmith CH, et al. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833-1840. [9] Fries JF, Spitz spitz Any of several northern dogs, including the chow chow, Pomeranian, and Samoyed, characterized by a dense, long coat, erect pointed ears, and a tail that curves over the back. In the U.S. P, Kraines RG, Holman HH. Measurement of patient outcome in arthritis. Arthritis Rheum. 1980;23:137-145. [10] Ramey DR, Raynauld J-P, Fries J. The Health Assessment Questionnaire 1992: status and review. Arthritis Care Arthritis Care is the UK's largest charity dedicated to supporting people with arthritis. The organisation is staffed and led by people who also have arthritis. It provides information and support on a range of issues related to living with arthritis. Research. 1992;5:119-129. [11] Gibbs J, Hughes S, Dunlop D, et al. Joint Impairment Impairment 1. A reduction in a company's stated capital. 2. The total capital that is less than the par value of the company's capital stock. Notes: 1. This is usually reduced because of poorly estimated losses or gains. 2. and ambulation in the elderly. J Am Geriatr 1993;41:1205-1211. [12] Falconer J, Hayes K. A simple method to measure gait for use in arthritis clinical research. Arthritis Care Research. 1991;4:52-57. [13] Domholdt E. Physical Therapy Research. Principles and Applications. Philadelphia, Pa: WB Saunders Co; 1993:154-155. [14] Heussler J. The Repeatability of Gait Parameters in Asymptomatic .Subjects Using an Electrical Footswitch System. Brisbane, Queensland, Australia: The University of Queensland The University of Queensland (UQ) is the longest-established university in the state of Queensland, Australia, a member of Australia's Group of Eight, and the Sandstone Universities. It is also a founding member of the international Universitas 21 organisation. ; 1993. Master's thesis. [15] Marks R. Reliability and validity of self-paced walking time measures for knee osteoarthritis. Arthritis Care Research. 1994;7:50-53. [16] Altman R, Asch E, Bloch D, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum. 1986;29:1039-1049. [17] Larsson L-E, Oldenrick P, Sandlund B, et al. The phases of the stride and their interaction in human gait. Scand J Rehabil Med. 1980;12:107-112. [18] Hirokawa S. Normal gait characteristics under temporal and distance constraints. Biomed Eng. 1989;11:449-456. [19] Bellamy N, Sothern R, Campbell J. Rhythmic rhyth·mic also rhyth·mi·cal adj. Of, relating to, or having rhythm; recurring with measured regularity. rhyth  mi·cal·ly adv. variations in pain
perception in osteoarthritis of the knee. J Rheumatol. 1990;17:364-372.[20] Norusis M. SPSS/PC[+ or -] [TM] 4.0 Base Manual. Chicago, III: SPSS Inc; 1990. [21] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater rat·er n. 1. One that rates, especially one that establishes a rating. 2. One having an indicated rank or rating. Often used in combination: a third-rater; a first-rater. reliability. Psychol Bull. 1979;86:420-428. [22] Fransen M, Heussler J, Margiotta E, Edmonds J. Quantitative gait analysis: comparison of rheumatoid rheumatoid /rheu·ma·toid/ (roo´mah-toid) 1. resembling rheumatism. 2. associated with rheumatoid arthritis. rheu·ma·toid adj. 1. Of or resembling rheumatism. arthritic and non-arthritic subjects. Australian Journal of Physiotherapy physiotherapy: see physical therapy. . 1994;40:191-199. [23] Fairfax A, Balnave R, Adams R. Variability of grip strength Grip strength is the force applied by the hand to pull on or suspend from objects. Optimum-sized objects permit the hand to wrap around a cylindrical shape with a diameter from one to three inches. during isometric isometric /iso·met·ric/ (-met´rik) maintaining, or pertaining to, the same measure of length; of equal dimensions. i·so·met·ric adj. 1. contraction. Ergonomics ergonomics, the engineering science concerned with the physical and psychological relationship between machines and the people who use them. The ergonomicist takes an empirical approach to the study of human-machine interactions. . 1995;38:1819-1830. [24] Hanke T, Pai Y-C, Rogers M. Reliability of measurements of body center-of-mass momentum during sit-to-stand in healthy adults. Phys Ther. 1995;75:105-113. M Fransen, DipPhy, MAPA MAPA Malaysia Airlines Pilots' Association MAPA Mexican-American Political Association MAPA Manila Action Plan for APEC MAPA Metropolitan Area Planning Agency MAPA Mine Action Program for Afghanistan (UN) , is Research Physiotherapist physiotherapist /phys·io·ther·a·pist/ (-ther´ah-pist) physical therapist. physiotherapist physical therapist. , Department of Rheumatology, The St George Hospital, Gray St, Kogarah, New South Wales, Australia 2217 (m.fransen@unsw.edu.au). Address all correspondence to Ms Fransen. J Crosbie, PhD, GradDipPhys, DipTP, is Associate Professor, School of Physiotherapy School of Physiotherapy is located in Lahore, Punjab, Pakistan. It is located in Mayo Hospital and is affiliated with King Edward Medical College. , Faculty of Health Sciences, The University of Sydney, Sydney, New South Wales New South Wales, state (1991 pop. 5,164,549), 309,443 sq mi (801,457 sq km), SE Australia. It is bounded on the E by the Pacific Ocean. Sydney is the capital. The other principal urban centers are Newcastle, Wagga Wagga, Lismore, Wollongong, and Broken Hill. , Australia. J Edmonds, MBBS MBBS, MBChB n abbr (BRIT) (= Bachelor of Medicine and Surgery) → título universitario MBBS, MBChB n abbr (Brit) (= Bachelor of Medicine and Surgery) → , FRACP FRACP Fellow of the Royal Australasian College of Physicians , is Director of Rheumatology, The St George Hospital, and Conjoint con·joint adj. 1. Joined together; combined: "social order and prosperity, the conjoint aims of government" John K. Fairbank. 2. Professor of Rheumatology, University of New South Wales The University of New South Wales, also known as UNSW or colloquially as New South, is a university situated in Kensington, a suburb in Sydney, New South Wales, Australia. , Kensington, New South Wales Kensington is a suburb in south-eastern Sydney, in the state of New South Wales, Australia. Kensington is located 6 kilometres south east of the Sydney central business district, in the local government area of the City of Randwick, in the Eastern Suburbs region. , Australia. The study was approved by the Southern Sydney Southern Sydney is a general term which is used to describe the southern metropolitan area of Sydney, in the state of new South Wales, Australia. Depending on the context, 'Southern Sydney' would include the suburbs in the local government areas of Rockdale, Kogarah and Area Health Service Ethics Committee ethics committee A multidisciplinary hospital body composed of a broad spectrum of personnel–eg, physicians, nurses, social workers, priests, and others, which addresses the moral and ethical issues within the hospital. See DNR, Institutional review board. . This research was presented as a podium podium In architecture, a pedestal on a large scale. It may be any of various elements that form the base of a structure, such as the platform forming the floor and substructure of a Classical temple, a low wall supporting columns, or the structurally or decoratively presentation at the national meeting of the American College of Rheumatology; October 22-27, 1995; San Francisco San Francisco (săn frănsĭs`kō), city (1990 pop. 723,959), coextensive with San Francisco co., W Calif., on the tip of a peninsula between the Pacific Ocean and San Francisco Bay, which are connected by the strait known as the Golden , Calif. |
|
||||||||||||||||
Printer friendly
Cite/link
Email
Feedback
Reader Opinion