Elsevier

Gait & Posture

Volume 52, February 2017, Pages 194-201
Gait & Posture

Full length article
The relative and absolute reliability of center of pressure trajectory during gait initiation in older adults

https://doi.org/10.1016/j.gaitpost.2016.11.042Get rights and content

Highlights

  • With regard to parameter, one to five trials should be averaged to ensure excellent reliability.

  • The ML and AP COP displacement in locomotor phase may be more suitable to discriminate between subjects.

  • The ML and AP median frequency in locomotor phase may be more appropriate to detect treatment-induced changes over time.

Abstract

It has been thought that for scientific acceptance of a parameter, its psychometric properties such as reliability, validity and responsiveness have critical roles. Therefore, this study was conducted to estimate how many trials are required to obtain a reliable center of pressure (COP) parameter during gait initiation (GI) and to investigate the effect of number of trials on the relative and absolute reliability. Twenty older adults participated in the study. Subjects began stepping over the force platform in response to an auditory stimulus. Ten trials were collected in one session. The displacement, velocity, mean and median frequency of the COP in the mediolateral (ML) and anteroposterior (AP) directions were evaluated. Relative reliability was determined using the intraclass correlation coefficient (ICC), and absolute reliability was evaluated using the standard error of measurement (SEM) and minimal detectable change (MDC95). The results revealed with respect to parameter, one to five trials should be averaged to ensure excellent reliability. Moreover, ICC, SEM% and MDC95% values were between 0.39–0.89, 4.84–41.5% and 13.4–115% for single trial and 0.86–0.99, 1.74–19.7% and 4.83–54.7% for ten trials averaged, respectively. Moreover, the ML and AP COP displacement in locomotor phase had the most relative reliability as well as the ML and AP median frequency in locomotor phase had the most absolute reliability. In general, the results showed that the COP-related parameters in time and frequency domains, based on average of five trials, provide reliable outcome measures for evaluation of dynamic postural control in older adults.

Introduction

Gait initiation (GI) is a daily living task consisting of three phases: postural phase [called Anticipatory Postural Adjustments (APAs)], weight transition, and locomotor phase [1], [2]. APAs are required to shift the center of pressure (COP) backward and toward the swing limb prior to the initial heel-off. This process provides the momentum required for efficient forward progression and allows the swing limb to leave the ground [3]. Impairment in function of the APA leads to a poor GI performance. Studies have suggested that the reduction of backward COP displacement is related to the risk of falling [4], [5]. In addition, the reduction of COP displacement toward the swing limb decreases the shift of center of mass (COM) toward the stance limb; in turn, the mediolateral stability during GI is decreased [6]. On the other hand, studies have reported that falls occur most frequently during the transitional phases of gait like initiation, termination or turning [5], [7]. Falls are a serious threat for older adults aged >65 years and result in injury or even death. Therefore, it is thought that the COP parameters derived from force platform during GI might be optimal measures for diagnosing postural control deficiencies, discriminating fallers from non-fallers, and assessing the effectiveness of balance training programs [4], [5], [8], [9], [10], [11]. However, it is generally admitted that before selecting a parameter as a standard measure, determining its precision and consistency across repeated measures is an essential prerequisite. The precision and consistency of a parameter refer to the repeatability of that variable and can be investigated using reliability methods [12]. COP parameters are subject to variability with two potential sources, instrument and biological phenomena being measured [13]. Therefore, a reliability study of these parameters is necessary to distinguish actual changes from the errors involved in what is being measured [14]. The presence of the measurement error is a major concern for clinicians to plan treatment, monitor progression, and assess the efficacy of treatment. Moreover, an insufficient number of trials may influence the reliability of COP parameters [15]. Although increasing the number of trials enhances the reliability, the time imposed and the fatigue on the participant or clinician must also be considered [16]. In GI studies, a broad range of trials (from 3 up to 10) have been used [4], [5], [8], [9], [10], [11]. Thus, it is important to specify the optimum number of trials required to achieve a reliable parameter. Despite the numerous studies that have utilized the COP parameters for differentiation between healthy and unhealthy subjects or the investigation of the effects of treatment [5], [8], [9], [10], [11], to the best of our knowledge, no study has evaluated their reliability during GI. It is worth mentioning that only Melzer and Halvarsson have investigated the reliability of voluntary step execution. However, these studies have focused on the temporal events and phases of voluntary step execution and not on COP parameters [17], [18]. Moreover, Sousa et al., have assessed the reliability of two methods (baseline-based method and maximal displacement-based method), for identifying the postural phase of GI in healthy and post-stroke subjects [6]. Therefore, this study has been conducted to estimate how many trials are required to obtain a reliable COP parameter during GI and to investigate the effect of the number of trials on the relative and absolute reliability.

Section snippets

Subjects

Twenty healthy older adults participated in the study (Table 1). The inclusion criteria were the following: Age  65 years old; Berg Balance Scale (BBS) score >40 that indicates subjects were in a low fall risk [5]; Timed Up & Go test (TUG) score ≤20 s that indicates subjects were mostly independent [19]; Activities-specific Balance Confidence Scale (ABC) score ≥50% that indicates subjects were in a moderate to high level of physical functioning [20]; Mini-Mental State Examination (MMSE) score ≥24

Results

The mean and standard deviation of COP parameters for each trial are reported in Table 2. There was no significant difference between trials for any COP parameter, which reflects the absence of any systematic bias (p > 0.05).

Table 3 represents the number of trials required for ICC ≥0.75, ICC and its 95% confidence interval (CI), SEM, SEM%, MDC95 and MDC95% obtained by single trial and 10 trials averaged.

Results demonstrated that, with respect to parameter, one to five trials should be performed

Discussion

The results revealed, with respect to parameter, one to five trials should be performed to ensure excellent reliability and also relative and absolute reliability tend to be improved as the number of trials increases. Some authors believe the AP displacement and AP velocity of COP in anticipatory phase are parameters which commonly demonstrate the age-related changes [4], [5], [26]. The present study estimated that for these two parameters at least four and five trials, respectively, are

Conflict of interest

None.

Acknowledgment

This study was supported by biomechanics laboratory of rehabilitation faculty, Tehran University of Medical Sciences.

References (29)

  • D. Lafond et al.

    Intrasession reliability of center of pressure measures of postural steadiness in healthy elderly people

    Arch. Phys. Med. Rehabil.

    (2004)
  • N. Cau et al.

    Center of pressure displacements during gait initiation in individuals with obesity

    J. Neuroeng. Rehabil.

    (2014)
  • K. Uemura et al.

    Effect of dual-tasking on the center of pressure trajectory at gait initiation in elderly fallers and non-fallers

    Aging Clin. Exp. Res.

    (2012)
  • A.S. Sousa et al.

    Reliability of two methods for identifying the postural phase of gait initiation in healthy and post-stroke subjects

    J. Appl. Biomech.

    (2015)
  • Cited by (7)

    • A systematic review of center of pressure measures to quantify gait changes in older adults

      2021, Experimental Gerontology
      Citation Excerpt :

      Ahmadi et al. (2018) also measured complexity of the COP using quantized dynamical entropy and sample entropy. Pressure mats (Chiu et al., 2013; Fuchioka et al., 2015; Hagedorn et al., 2013; Lu et al., 2017; Ryu et al., 2019; Sole et al., 2017), pressure-sensing insoles (Howcroft et al., 2016; Howcroft et al., 2018; Li et al., 2019; Park et al., 2013) and force plates (Hass et al., 2004; Hayati et al., 2018; Khanmohammadi et al., 2015; Khanmohammadi et al., 2017a; Khanmohammadi et al., 2017b; Kim, 2009a, 2009b; Kim et al., 2012; Kim et al., 2013; Lelard et al., 2017; Polcyn et al., 1998; Spencer and van der Meer, 2012; Svoboda et al., 2017; Uemura et al., 2011; Uemura, 2012a; Uemura, 2012c; Uemura, 2012b; Vallabhajosula et al., 2014; van Andel et al., 2019; van Kooten et al., 2018; Wang et al., 2015; You et al., 2009) were used to measure COP trajectory during walking with force plates the most common (n = 21). One study also adopted Nintendo Wii balance board (Nintendo, Kyoto, Japan) to record COP trajectory during the gait initiation task (Lee et al., 2019).

    View all citing articles on Scopus
    View full text