Applying the Minimal Detectable Change of a Static and Dynamic Balance Test Using a Portable Stabilometric Platform to Individually Assess Patients with Balance Disorders

healthcare Article Applying the Minimal Detectable Change of a Static and Dynamic Balance Test Using a Portable Stabilometric Platform to Individually Assess Patients with Balance Disorders Juan De la Torre 1,* , Javier Marin 1, Marco Polo 2and JoséJ. Marín1,3 1 IDERGO-Research and Development in Ergonomics, Biomechanical Laboratory, I3A-University Institute of Research of Engineering of Aragon, University of Zaragoza, 50009 Zaragoza, Spain; [email protected] (J.M.); [email protected] (J.J.M.) 2 Physical Medicine and Rehabilitation, Hospital of Alcañiz, 44600 Teruel, Spain; mar[email protected] 3Department of Design and Manufacturing Engineering, University of Zaragoza, 50009 Zaragoza, Spain *Correspondence: [email protected] Received: 22 July 2020; Accepted: 12 October 2020; Published: 14 October 2020   Abstract: Balance disorders have a high prevalence among elderly people in developed countries, and falls resulting from balance disorders involve high healthcare costs. Therefore, tools and indicators are necessary to assess the response to treatments. Therefore, the aim of this study is to detect relevant changes through minimal detectable change (MDC) values in patients with balance disorders, specifically with vertigo. A test-retest of a static and dynamic balance test was conducted on 34 healthy young volunteer subjects using a portable stabilometric platform. Afterwards, in order to show the MDC applicability, eight patients diagnosed with balance disorders characterized by vertigo of vestibular origin performed the balance test before and after a treatment, contrasting the results with the assessment by a specialist physician. The balance test consisted of four tasks from the Romberg test for static balance control, assessing dynamic postural balance through the limits of stability (LOS). The results obtained in the test-retest show the reproducibility of the system as being similar to or better than those found in the literature. Regarding the static balance variables with the lowest MDC value, we highlight the average velocity of the center of pressure (COP) in all tasks and the root mean square (RMS), the area, and the mediolateral displacement in soft surface, with eyes closed . In LOS, all COP limits and the average speed of the COP and RMS were highlighted. Of the eight patients assessed, an agreement between the specialist physician and the balance test results exists in six of them, and for two of the patients, the specialist physician reported no progression, whereas the balance test showed worsening. Patients showed changes that exceeded the MDC values, and these changes were correlated with the results reported by the specialist physician. We conclude that (at least for these eight patients) certain variables were sufficiently sensitive to detect changes linked to balance progression. This is intended to improve decision making and individualized patient monitoring. Keywords: posturography; vertigo; vestibular disorders; test-retest; patient-level analysis 1. Introduction Balance disorders have a high prevalence among elderly people in developed countries [ 1 ]. Connected with the current trend of increasing age of the population [ 2 , 3 ], this has led to a rise in pathologies that affect balance, causing an increased risk of falls in the elderly population. Around 30% of people aged over 65 years, and more than 50% of individuals in health centers or care homes suffer Healthcare 2020,8, 402; doi:10.3390/healthcare8040402 www.mdpi.com/journal/healthcare Healthcare 2020,8, 402 2 of 18 one or more falls per year, and between approximately 20% and 30% of the global population has had or will have a vertiginous episode in their lives [4–10]. Therefore, it is necessary to have tools and indicators to assess the response to treatments of pathologies that affect the sensory systems involved in balance: visual, vestibular, and proprioceptive [ 11 ]. The effects of a given treatment on a patient through the performance of previous and subsequent tests can help in making decisions about adjusting, changing, or stopping the treatment [12]. Stabilometric platforms are useful in assessing balance because they obtain numerous parameters of the centers or pressure (COP) [ 13 , 14 ]. Stabilometric platforms can assess static balance control and dynamic postural balance through different variables and application methods [ 15 , 16 ]. On the one hand, in static balance control methods (such as the Romberg test), subjects must maintain their COP within the support base throughout the assessment period of time. On the other hand, the assessment of dynamic postural balance, which is vital for motor control, involves measuring the limits of stability (LOS), corresponding to the maximum voluntary angle or distance in which an individual can regulate their COP in a given direction without losing balance [ 17 ]. Stabilometric platforms can obtain objective information related to balance pathologies in clinical practice to improve the quality of healthcare and the provided treatments [4,18–20]. Therefore, a balance test would be beneficial for clinicians by providing them with an objective assessment device; however, the integration of this type of test in clinical practice involves various difficulties and constraints. Although stabilometric platforms are normally used in the lab, their implementation in the clinic is complex due to difficulty of use and manageability, as well as size, which may call into question clinical applicability [ 21 ]. The protocol must be validated and adapted for continuous and assiduous use in the clinic [ 21 , 22 ]. Therefore, it is relevant to conduct studies to measure the degree of variability of a balance test to achieve a successful clinical application. The minimal detectable change (MDC) index represents the variability of the measures of each variable, i.e., the consistency of the variables, which includes different aspects: the variability of the instrument, the inherent variability of the person and their learning (test duration, visual condition, or position of the feet) and the procedures and protocols applied to perform the test. Therefore, if we obtain a change of one variable that is higher than its MDC value, this would be a relevant change, which could be caused by the treatment effect and not by the test variability [23,24]. No balance assessment studies, based on the MDC results of the test-retest, to detect changes in the results associated with balance progression have been conducted for patients with balance disorders. Numerous test-retest studies have been found in the COP measures but have not included patients with balance disorders in the application [4,19,24–32]. Therefore, the aim of this study is to detect relevant changes through MDC values in patients with balance disorders, specifically with vertigo. To address this, the following research actions/steps are proposed: (1) to perform a test-retest of a static and dynamic balance test in a sample of healthy subjects identifying the most sensible variables with the lower MDC values; (2) to analyze whether relevant changes are detected in the results of the balance test before and after the application of a treatment in eight patients diagnosed with balance disorders with vertigo of vestibular origin, contrasting the results with the progression observed by a specialist physician. 2. Materials and Methods 2.1. Participants and Ethics Statement A test-retest study, which consists of repeating the test at two different times with a homogeneous sample of participants under the same conditions, of a static and dynamic balance test was conducted on a sample of 34 healthy young volunteer subjects. Participating in this study were 20 males and 14 females, (age 22.89 ± 3.51 years, height 172.51 ± 9.01 cm, weight 67.38 ± 11.82 kg, body fat index 20.07 ± 9.12 %, foot length [33] 25.39 ±1.81 cm, abdominal perimeter 79.76 ±9.77 cm). Healthcare 2020,8, 402 3 of 18 The calculation for the choice of 34 subjects was based on the research of Bujang et al. [ 34 ], an article in which the relationship between the interclass correlation coefficient (ICC), statistical power, and the number of subjects is established. Taking into account the necessity for two measurements to be made per subject, in order to satisfy the requirement for repetition of the test at two different times, to set a statistical power of 80% and to establish a minimum ICC of 0.5, the sample for analysis needed to number at least 22. Therefore, the choice of 34 subjects is considered appropriate. The inclusion criteria established for participation in the study were (i) no history of neurological, visual, or vestibular alterations, (ii) no record of musculoskeletal or neurological diseases in the last year, and (iii) no history of limb surgery that may affect the patient’s balance. It has been demonstrated in several studies that the variability of variables related to the COP are higher in young subjects, which has motivated the selection of the sample [4,28]. As shown in Section 2.5, eight patients with balance disorders characterized by vertigo of vestibular origin were also analysed to apply and assess the test-retest results. The patients were analysed before and after treatment to obtain the individual results of their progression. The choice of eight patients was based on the sample size calculation for quantitative variables of Charan et al. [35]. The formula used was the following: N=(Z1−α/2)2×SD2/d2(1) where Z 1−α/2 is the standard normal variate with p v <0.05 (error type 1), corresponding to value of 1.96 (in the majority of studies p v values are considered significant below 0.05, hence 1.96 is used in the formula); SD is the standard deviation of the variable (standard deviation value can be taken from previously studies); and d is the absolute error or precision (decided by researcher). The SD has been selected from the study conducted by Balaguer et al. [ 20 ], for using variables and tests similar to those used in this study. Specifically, the test in which all the sensory systems are available has been considered (rigid surface with eyes open), selecting one of the most used variables in balance studies [ 29 , 36 ]: the average speed of the COP. The SD value, for this variable and for this test is 6 mm/s; selecting a d value of 4.1 mm/s, and maintaining a Z 1−α/2 value of 1.96, we obtained an Nvalue of 8.22 in the eight selected patients. The selected patients met the following inclusion criteria: (i) between 65 and 75 years old and (ii) having suffered a vertiginous episode in the last year. The following were the exclusion criteria: (i) presented acute osteo-muscular pathology in the lower limbs or lumbar spine, which may alter the outcome of the stabilometric platform, (ii) presented any amputation in the lower limbs, (iii) presented oncological pathology or was in active treatment with chemotherapy, radiotherapy, or hormonal therapy, (iv) presented degenerative diseases (which can cause loss of capacity during the assessment period of time), (v) having been influenced by work or family events ( or other kind of events ) that have detrimentally affected the patient’s mood. The study was approved by the Government of Aragon’s Human Research Ethics Committee (CEICA) (16 January 2019). Prior to the beginning of the tests, the participants signed a form, consenting to undergo them and indicating that they understood the aim of the study. 2.2. Instrumentation The device used was the stabilometric platform MoveHuman-Dyna UZ, which was designed and manufactured by the IDERGO (Research and Development in Ergonomics, University of Zaragoza, Spain ) research group (Figure 1). This is a static posturography device designed for research, which comprises four load cells and a lightweight aluminum structure, whose dimensions and characteristics are detailed in the study of [ 18 ]. The findings of this device can be replicated in a straightforward manner by other researchers, which enhances the applicability of this study. The acquisition and processing of the platform data, as well as the format and method of exporting them, have been carried out according to the procedure used by [ 18 ]. Processing the force Healthcare 2020,8, 402 4 of 18 data as a function of the cells’ position means we can calculate the real-time position of the trajectory that describes the position of the CoP by applying the appropriate formula [7,13]. Healthcare 2020, 8, x 4 of 18 dimensions, resolution, sampling, etc. The precision parameters (accuracy, precision, linearity, dimensions and resolution) were obtained through an experiment in which the metrological characteristics of the platform were tested with a gold standard force platform, along with the error of measurement [17]. Specifically, the gold standard force platform with which it was compared was the AMTI 0R6-7-1000, U.S. Patent Number 4.493.220. The two platforms were positioned one above the other, in order to enable their results to be directly compared and examined for any variability, according to the study of Huurnink et al. (2013) [38]. The stabilometric platform has since been used in several research projects with patients in different hospitals, both public and private; all these research projects have been approved by the CEICA Committee. The use of this stabilometric platform in this study was justified by its portability characteristics, dimensions, and weight that enable its use by physicians in an examination room, often with limited space [13,18,37,39,40]. A portable device such this platform allows its use in different medical centers due to the quicker installation process and smaller space required, which improves its accessibility and applicability [41–43]. Figure 1. Stabilometric platform and environmental test condition. 2.3. Protocol The static and dynamic balance were both assessed with a set of tests previously applied in other studies [18]. The static balance control was assessed with a test based on the Romberg test and the Modified Clinical Test of Sensory Interaction in Balance (CTSIB-M), with consideration given to four different situations: (1) rigid surface with eyes open (RSEO), (2) rigid surface with eyes closed (RSEC), (3) soft surface with eyes open (SSEO), and (4) soft surface with eyes closed (SSEC). On the other hand, the dynamic postural balance was assessed measuring the LOS that a subject is able to reach and, with it, the management capacity of COP [17]. The dynamic LOS test was based on protocols found in the literature [44]. The inclusion of the LOS, complementary to the assessment of the static balance control, provides additional value to the balance assessment protocol [4,8,45] The protocol applied in the tests (the position of the body, arms and feet during the test [18], environmental conditions (e.g. noise, space, etc.) and the additional instrumentation used as a foam rubber for soft surface, instruments for anthropometric data collection, etc.) is that used by Delatorre et al (2017) for this stabilometric platform (Figure 1). This protocol fulfils certain clinical conditions [21,46–48]; it must be fast and should not require multiple repetitions to issue a definite, consistent result [21]. It must also be effective, with clear, specific stages defined by the instructions given by the operator that are understandable by the patient [28]. The test-retest in the sample of healthy subjects was performed on the same day with an interval of 6 hours between the two trials. In this interval, the participants did not perform physical activities that required extra muscular activation and/or caused fatigue that could affect the results of the Figure 1. Stabilometric platform and environmental test condition. Likewise, in accordance with the aforementioned study, the stabilometric platform “meets the standards established by the International Society for Posture and Gait Research (ISPGR) for its clinical application” [ 37 ] in relation to various parameters, such as accuracy, precision, linearity, dimensions, resolution, sampling, etc. The precision parameters (accuracy, precision, linearity, dimensions and resolution) were obtained through an experiment in which the metrological characteristics of the platform were tested with a gold standard force platform, along with the error of measurement [ 17 ]. Specifically, the gold standard force platform with which it was compared was the AMTI 0R6-7-1000, U.S. Patent Number 4.493.220. The two platforms were positioned one above the other, in order to enable their results to be directly compared and examined for any variability, according to the study of Huurnink et al. (2013) [ 38 ]. The stabilometric platform has since been used in several research projects with patients in different hospitals, both public and private; all these research projects have been approved by the CEICA Committee. The use of this stabilometric platform in this study was justified by its portability characteristics, dimensions, and weight that enable its use by physicians in an examination room, often with limited space [ 13 , 18 , 37 , 39 , 40 ]. A portable device such this platform allows its use in different medical centers due to the quicker installation process and smaller space required, which improves its accessibility and applicability [41–43]. 2.3. Protocol The static and dynamic balance were both assessed with a set of tests previously applied in other studies [ 18 ]. The static balance control was assessed with a test based on the Romberg test and the Modified Clinical Test of Sensory Interaction in Balance (CTSIB-M), with consideration given to four different situations: (1) rigid surface with eyes open (RSEO), (2) rigid surface with eyes closed (RSEC), (3) soft surface with eyes open (SSEO), and (4) soft surface with eyes closed (SSEC). On the other hand, the dynamic postural balance was assessed measuring the LOS that a subject is able to reach and, with it, the management capacity of COP [ 17 ]. The dynamic LOS test was based on protocols found in the literature [ 44 ]. The inclusion of the LOS, complementary to the assessment of the static balance control, provides additional value to the balance assessment protocol [4,8,45] The protocol applied in the tests (the position of the body, arms and feet during the test [ 18 ], environmental conditions (e.g., noise, space, etc.) and the additional instrumentation used as a Healthcare 2020,8, 402 5 of 18 foam rubber for soft surface, instruments for anthropometric data collection, etc.) is that used by Delatorre et al (2017) for this stabilometric platform (Figure 1). This protocol fulfils certain clinical conditions [ 21 , 46 – 48 ]; it must be fast and should not require multiple repetitions to issue a definite, consistent result [ 21 ]. It must also be effective, with clear, specific stages defined by the instructions given by the operator that are understandable by the patient [28]. The test-retest in the sample of healthy subjects was performed on the same day with an interval of 6 hours between the two trials. In this interval, the participants did not perform physical activities that required extra muscular activation and/or caused fatigue that could affect the results of the second trial. The tests of all study participants were coordinated by the same operator. The time interval between the tests performed by the patients was 3 months (process detailed in Section 2.5). 2.4. Statistical Analysis: Test-Retest Study 2.4.1. Selection of Variables The variables selected for the present test-retest study were those determined by [ 18 ] to be more significant in balance assessment studies, whose details and methods are also explained in the same study. The variables selected for the assessment of static and dynamic balance were the range of displacement in the anteroposterior and mediolateral directions, area (surface area covered by the trajectory of the COP), average speed of the COP, and root mean square (RMS) position. Additionally, in the LOS test, two more variables were assessed: the COP limits (maximum displacement reached along each axis of the octagon radii), and the “success” variable (quantification of the management and coordination of the COP along each axis of the octagon radii), both defined in a previous study [18]. 2.4.2. Minimal Detectable Change Calculation The MDC index represents the variability of the measures of each variable. If a change of one variable that is lower than its MDC value is detected, it would not be considered relevant, since it is lower than the variability of the test. Therefore, to narrow the intrinsic variability of the test, we considered that the best choice would be to perform the test-retest with healthy and young people (this topic will be justified in the discussion section). The MDC index was calculated for each of the variables resulting from the test-retest study. The value of the MDC was calculated from the following Equations (2) and (3) [27,30,49–53]: MDC95% =1.96 ×√2 SEM (2) SEM =SD-pooled ×√(1 −r) (3) where r is the interclass correlation coefficient (ICC), SD-pooled is the pooled average of the standard deviation of the test and retest, and the SEM is the standard deviation error of measurement. The dimensionless value of the effect size (MDC.es 95%) was also calculated using Equation (4), which indicates the number of standard deviations that the experiment is capable of detecting [54]: MDC.es 95% =MDC95% /SD test (4) where SD test is the standard deviation of the test (initial test). The ICC results can be classified according to Cicchetti (1994) [ 55 ], who provided the following intervals to characterize the ICC inter-rater agreement measures: below 0.40: poor; between 0.40 and 0.59: fair; between 0.60 and 0.74: good; between 0.75 and 1.00: excellent. 2.5. Clinical Application Foundations: Assessment of Patients’ Progression To verify whether the variables in the balance test could detect changes that exceed the MDC index before and after treatment, eight patients diagnosed with balance disorders characterized by Healthcare 2020,8, 402 6 of 18 vertigo of vestibular origin performed the aforementioned test. This allowed us to verify whether the analyses of the results of the pre-tests and post-tests coincided with the subjective patient progression (positive, null or negative progression) perceived by a specialist physician. 2.5.1. Patients’ Initial Diagnosis The patients were referred by the Otorhinolaryngology Service of the Alcañiz Hospital ( Teruel, Spain ) after being diagnosed with a balance disorder. The methods for determining the vestibular deficit were medical history, magnetic resonance imaging, videonystagmography, and tests such as the Dix-Hallpike maneuver (Table 1). Table 1. Data of patients assessed. Patient Code Age Gender Anamnesis and Examination 01 72 Male BPPV with associated signs of bilateral hearing loss (detected by audiometry) and nystagmus. 02 73 Female Ménière syndrome. 03 70 Male BPPV with associated signs of bilateral hearing loss (detected by audiometry). 04 73 Female Osteosclerosis. Bilateral hearing loss. 05 68 Male Ménière syndrome. Unilateral tinnitus and pathological nystagmus. 06 74 Male Ménière syndrome. Unilateral tinnitus. 07 68 Female Vestibular hypofunction. Pathological nystagmus. 08 75 Male BPPV with associated signs of bilateral hearing loss (detected by audiometry). BPPV: Benign Paroxysmal Peripheral Vertigo. 2.5.2. Clinician 1 Assessment: History and Physical Examination A doctor (clinician 1) from the Physical Medicine and Rehabilitation Department of the Alcañiz Hospital (Teruel, Spain) evaluated the patients using medical history and a physical examination, as well as functional balance assessment tests such as the Unterberger test [ 56 , 57 ], the up and go test [ 58 , 59 ], and unipodal support test [ 60 ]. Clinician 1 prescribed the rehabilitation treatment, which consists of a set of vestibular rehabilitation exercises (to be performed by the patients), which is commonly used in the clinic, establishing these over a period of three months [ 61 , 62 ]. Clinician 1 evaluated the patients at two different times, once before starting the treatment (pre-data), and once three months after starting the treatment (post-data). 2.5.3. Clinician 2 Assessment: Patient Progression Evaluation The preand postdata collected by clinician 1 were assessed by a specialist physician ( clinician 2 ), which allowed an assessment of the balance progression of each of the eight patients. To avoid the results being influenced or contaminated by the interaction between the clinicians, there was no contact between them during the research. The assessment of clinician 2 established three possible categories to evaluate patient progression: positive, null or negative progression. 2.5.4. Magnitude-Based Decision (MBD) to Monitor Patients with Balance Disorders To measure the effects of a treatment on a specific patient, it is necessary to evaluate the treatment status at different times. Thus, the patient-level approach proposed by Hopkings (2017) [ 63 ] was applied to obtain personalized results for each patient. This approach allows the assessment the change between two measurements in an individual through the magnitude-based decision (MBD) method (formerly known as magnitude-based inferences) [64]. Hopkings (2017) [ 63 ] introduced the patient-level approach providing “A Spreadsheet for Monitoring an Individual’s Changes”. Based on this spreadsheet, we developed a script that applies the MBD method using as input the measurements taken from the preand post-balance tests of one specific patient, and the threshold MDC95% previously calculated in the test-retest study. The script was developed using WorldViz-Vizard 6.2 (based on Python 2.7), and the Pandas and Matplotlib libraries. Healthcare 2020,8, 402 7 of 18 The patients in the study performed the Romberg and LOS tests in the rehabilitation clinic of the Alcañiz Hospital on two occasions, once before starting the treatment and another three months after starting the treatment. Using the script that we have developed, it is possible to measure how much each variable has changed and whether the change detected is significant. According to the MBD method, inputs are required to analyze each variable: • Xdif: difference between the measures taken at two temporal points: pre-value and post-value (Equation (5)). In this case, the pre-value is the measure of each variable just before starting the treatment; and the post-value is the measure three months after starting the treatment. Xdi f =Xpost −Xpre (5) • MBD threshold: for this method, a threshold (numerical value) must be defined from which a change is considered relevant. In our case, we selected the MDC previously calculated. In this approach, the statistical analysis detects whether the changes exceed a particular threshold, in our case the MDC value. Specifically, we determined where the confidence interval of the difference was located (between preand post-tests) in relation to the thresholds of the MDC [ 54 , 65 ]. The intervals of the differences of each variable between the pre-test and post-test results were determined to be on the negative side of the MDC threshold (% negative differences), within the threshold ( % trivial differences ), or on the positive side (% positive differences), respectively [ 63 ]. We deal with the selection of the MDC as threshold in the discussion section. Firstly, the value and sign (positive or negative) of Xdif is obtained through the difference between the pre-value and post-value. Subsequently, following the calculation method set forth by [ 63 ], the probability of change is obtained, which can be defined as the probability that the difference between the two values is relevant. This probability corresponds to the percentage of the confidence interval of the difference (calculated using the Xdif) that is outside of the range (+MDC, − MDC). Finally, the probability that the change is relevant was qualitatively classified (as proposed by Batterham and Hopkins (2006)) [ 54 ]. The qualitative classification of the significance of the changes is: most unlikely (<1%), very unlikely (1% to 5%), unlikely (5% to 25%), possibly (25% to 75%), probably (75% to 95%), very likely (95% to 99%) and most likely (>99%). For the calculations and different graphs presented in this study, Python 2.7 and the numpy, scipy and pandas modules were used. 3. Results 3.1. Test-Retest Results from Balance Analysis. Minimal Detectable Changes Index Table 3shows the results of the test-retest of the variables selected. The values of the means ( µ ) and standard deviations (SD) of each of the analyzed variables are shown by task. Likewise, the results of the variability through ICC 3,k (similar to ICC 2,1 ) [ 24 ], absolute value of the MDC (95%), and dimensionless value of the effect size (MDC.es 95%) are included. Table 2. Test-retest results from balance tests. Minimal detectable changes index. Balance Tasks Variables Test µ(SD) Retest µ(SD) ICC MDC95_es MDC95 RSEO COP mean speed [mm/s] * 8.1 (2.3) 8.4 (2.2) 0.87 0.9 2.3 RMS [mm] 5.1 (1.2) 5.3 (1.7) 0.61 2.1 2.6 Area [cm2]1.7 (0.9) 1.7 (0.9) 0.71 1.5 1.4 AP disp. [mm] 19 (4.7) 19.5 (6.1) 0.42 2.4 11.6 ML disp. [mm] 15.8 (4.4) 15.8 (4.8) 0.70 1.5 7.1 RSEC COP mean speed [mm/s] * 13.9 (4.2) 13.6 (3.9) 0.92 0.7 3.3 RMS [mm] * 6.2 (1.9) 6.5 (1.8) 0.87 0.9 1.9 Area [cm2] * 3.5 (1.9) 3.4 (1.8) 0.85 1.0 2.0 AP disp. [mm] 24.3 (6.1) 25.9 (7.7) 0.75 1.5 9.7 ML disp. [mm] * 22.4 (8.3) 22.7 (7.6) 0.85 1.0 8.7 Healthcare 2020,8, 402 8 of 18 Table 2. Cont. Balance Tasks Variables Test µ(SD) Retest µ(SD) ICC MDC95_es MDC95 SSEO COP mean speed [mm/s] * 13.7 (2.7) 13.5 (2.3) 0.88 0.9 2.5 RMS [mm] 6.9 (1.3) 6.9 (1.5) 0.74 1.4 2.0 Area [cm2]3.7 (1.0) 3.4 (1.1) 0.76 0.4 1.5 AP disp. [mm] 27.5 (5.1) 27.2 (6.7) 0.72 1.7 8.8 ML disp. [mm] 23.4 (4.3) 22.2 (4) 0.56 1.7 7.7 SSEC COP mean speed [mm/s] * 36.2 (7.9) 34 (6.6) 0.83 1.0 8.5 RMS [mm] 13.6 (2.2) 13.3 (2.3) 0.77 1.3 3.0 Area [cm2]17.2 (5.3) 15.7 (4.7) 0.74 1.3 7.2 AP disp. mm] 54.8 (12.4) 52.7 (10.9) 0.61 1.6 20.4 ML disp. [mm] 51.6 (10.6) 50.4 (9.5) 0.42 2.0 21.4 LOS COP mean speed [mm/s] * 15.9 (3.1) 15.4 (3.0) 0.93 0.7 2.3 RMS [mm] * 5.8 (0.3) 5.9 (0.3) 0.96 0.5 0.2 Area [cm2] * 179.5 (43.7) 183.2 (45.2) 0.97 0.4 21.7 AP disp. [mm] 153.8 (20.4) 157 (19.7) 0.94 0.7 14.4 ML disp. [mm] 157.1 (20) 158 (23.1) 0.94 0.7 14.8 Lim.COP.Forward [mm] * 84.6 (17.5) 87.5 (15.3) 0.9 0.8 15.0 Lim.COP.Forward-rightward [mm] * 90.4 (14.4) 91.8 (11.8) 0.93 0.7 10.2 Lim.COP.Rightward [mm] 77.3 (12.2) 75.5 (12.2) 0.83 1.1 14.0 Lim.COP.Backward-rightward [mm] 68.2 (12.7) 68 (10.6) 0.88 0.8 11.4 Lim.COP.Backward [mm] 68.1 (14.3) 68.5 (13.6) 0.84 1.0 15.6 Lim.COP.Backward-leftward [mm] 71.2 (12.2) 71 (14.1) 0.85 1.1 14.4 Lim.COP.Leftward [mm] 77.6 (10.5) 81.4 (13.8) 0.86 1.2 13.1 Lim.COP.Forward-leftward [mm] * 91 (14.8) 92.6 (12.4) 0.89 0.8 12.5 Success.Forward [%] 75.7 (14.4) 76.8 (11.7) 0.5 1.7 25.8 Success.Forward-rightward [%] 79.4 (12.2) 79.1 (13.9) 0.87 1.0 13.2 Success.Rightward [%] 78.3 (12.3) 80.4 (10.7) 0.68 1.4 18.1 Success.Backward-rightward [%] 79.1 (11.6) 78 (9.9) 0.42 1.9 23.0 Success.Backward [%] 83.5 (11.) 85.6 (8.2) 0.57 1.5 18.5 Success.Backward-leftward [%] 80.4 (10.5) 82.8 (11.7) 0.2 2.6 27.7 Success.Leftward [%] 80.8 (11.7) 81.6 (11.1) 0.8 1.2 14.4 Success.Forward-leftward [%] 78.8 (13.3) 80.5 (11.4) 0.77 1.2 16.5 µ : mean; SD: standard deviation; ICC: intraclass correlation coefficient; MDC95_es: minimal detectable change in dimensionless value effect size at 95%.; MDC95: minimal detectable change in absolute value at 95%; RSEO: rigid surface, open eyes; RSEC: rigid surface, eyes close; SSEO: soft surface, eyes open; SSEC: soft surface, eyes close; LOS: limits of stability; COP: center of pressure; RMS: root mean square; AP: anteroposterior; ML: mediolateral. * Balance variables selected with higher ICC. 3.2. Results of the Patients-Level Study Due to the large amount of information, the results of the patient with Code 01 are presented in this section as an example, and the results of the remaining patients are presented in the Supplementary Material. Thus, for each of the eight patients, the same information as shown in Table 3for Patient 1 has been calculated, which includes the values of the (a) variables for the preand post-tests, (b) differences, (c) MDC, and (d) percentages of negative, trivial, and positive differences (-/0/+). In addition, Figures 2and 3show the produced changes (black lines) in relation to the MDC (light grey rectangle) in each variable for each of the tests. Table 3. Results of the patient 01. Balance Tasks Variables Value Pre Value Post Difference MDC % (−/0/ +) RSEO COP mean speed [mm/s] 13.3 11.2 −2.0 2.3 41/59/0 RMS [mm] 6.5 5.0 −1.5 2.6 22/78/0 Area [cm2]3.4 2.1 −1.2 1.4 43/57/0 AP disp. [mm] 20.0 17.0 −3.0 11.6 8/91/1 ML disp. [mm] 22.4 17.7 −4.7 7.1 26/74/0 RSEC COP mean speed [mm/s] 20.3 16.9 −3.4 3.3 53/47/0 RMS [mm] 7.6 7.5 −0.1 1.9 4/94/2 Area [cm2]5.1 3.3 −1.8 2.0 40/60/0 AP disp. [mm] 26.9 31.6 4.7 9.7 0/84/16 ML disp. [mm] 30.4 20.0 −10.4 8.7 65/35/0 SSEO COP mean speed [mm/s] 18.5 18.0 −0.6 2.5 7/92/1 RMS [mm] 7.3 6.8 −0.5 2.0 7/92/1 Area [cm2]4.9 3.7 −1.1 1.5 33/67/0 AP disp. [mm] 23.3 23.3 0.0 8.8 3/94/3 ML disp. [mm] 44.3 28.4 −15.9 7.7 98/2/0 Healthcare 2020,8, 402 9 of 18 Table 3. Cont. Balance Tasks Variables Value Pre Value Post Difference MDC % (−/0/ +) SSEC COP mean speed [mm/s] 62.6 49.4 −13.2 8.5 86/14/0 RMS [mm] 19.2 15.0 −4.2 3.0 78/22/0 Area [cm2]41.0 25.4 −15.6 7.2 99/1/0 AP disp. [mm] 102.4 62.2 −40.2 20.4 97/3/0 ML disp. [mm] 66.5 72.5 6.0 21.4 1/91/8 LOS COP mean speed [mm/s] 19.0 21.2 2.2 2.3 0/52/48 RMS [mm] 5.5 6.2 0.7 0.2 0/0/100 Area [cm2]170.0 234.0 64.0 21.7 0/0/100 AP disp. [mm] 154.9 176.8 21.9 14.4 0/16/84 ML disp. [mm] 150.3 179.5 29.2 14.8 0/3/97 Lim.COP.Forward [mm] 86.8 108.2 21.4 15.0 0/21/79 Lim.COP.Forward-rightward [mm] 9.5 108.3 15.8 10.2 0/14/86 Lim.COP.Rightward [mm] 0.0 92.8 92.8 14.0 0/0/100 Lim.COP.Backward-rightward [mm] 60.4 72.5 12.1 11.4 0/45/55 Lim.COP.Backward [mm] 32.7 69.0 36.4 15.6 0/1/99 Lim.COP.Backward-leftward [mm] 60.0 82.1 22.1 14.4 0/15/85 Lim.COP.Leftward [mm] 72.7 87.0 14.3 13.1 0/43/57 Lim.COP.Forward-leftward [mm] 106.6 104.7 −1.9 12.5 5/93/2 Success.Forward [%] 50.7 68.8 18.2 25.8 0/72/28 Success.Forward-rightward [%] 69.0 73.6 4.6 13.2 1/89/11 Success.Rightward [%] 43.9 88.2 44.2 18.1 0/0/100 Success.Backward-rightward [%] 45.3 81.5 36.2 23.0 0/13/87 Success.Backward [%] 53.9 91.9 37.9 18.5 0/2/98 Success.Backward-leftward [%] 43.6 70.2 26.6 27.7 0/53/47 Success.Leftward [%] 55.9 75.1 19.1 14.4 0/26/74 Success.Forward-leftward [%] 58.5 79.2 20.8 16.5 0/31/69 Healthcare 2020, 8, x 10 of 18 Figure 2. Patient level analysis. Romberg Test. Minimal detectable change (MDC): light grey bars; Differences: thin Black bars. Figure 3. Patient level analysis. Limits of Stability. Minimal detectable change: light grey bars; Differences: Black thin bars. Figure 2. Patient level analysis. Romberg Test. Minimal detectable change (MDC): light grey bars; Differences: thin Black bars. Healthcare 2020,8, 402 16 of 18 25. Geldhof, E.; Cardon, G.; De Bourdeaudhuij, I.; Danneels, L.; Coorevits, P.; Vanderstraeten, G.; De Clercq, D. Static and dynamic standing balance: Test-retest reliability and reference values in 9 to 10 year old children. Eur. J. Pediatr. 2006,165, 779–786. [CrossRef] 26. Lafond, D.; Corriveau, H.; H é bert, R.; Prince, F. Intrasession reliability of center of pressure measures of postural steadiness in healthy elderly people. Arch. Phys. Med. Rehabil. 2004,85, 896–901. [CrossRef] 27. Lee, P.; Liu, C.; Fan, C.; Lu, C.; Lu, W.; Hsieh, C. The test–retest reliability and the minimal detectable change of the Purdue Pegboard Test in schizophrenia. J. Formosan Med. Assoc. 2013,112, 332–337. [CrossRef] 28. Pagnacco, G.; Carrick, F.R.; Wright, C.H.G.; Oggero, E. Between-subjects differences of within-subject variability in repeated balance measures: Consequences on the minimum detectable change. Gait Posture 2015 , 41, 136–140. [CrossRef] 29. Pinsault, N.; Vuillerme, N. Test–retest reliability of centre of foot pressure measures to assess postural control during unperturbed stance. Med. Eng. Phys. 2009,31, 276–286. [CrossRef] 30. Steffen, T.; Seney, M. Test-retest reliability and minimal detectable change on balance and ambulation tests, the 36-item short-form health survey, and the unified Parkinson disease rating scale in people with parkinsonism. Phys. Ther. 2008,88, 733–746. [CrossRef] 31. Lee, H.; Granata, K.P. Process stationarity and reliability of trunk postural stability. Clin. Biomech. 2008 , 23, 735–742. [CrossRef] [PubMed] 32. De Oliveira Silva, D.; Briani, R.V.; Pazzinatto, M.F.; Ferrari, D.; Arag ã o, F.A.; de Albuquerque, C.E.; Alves, N.; de Azevedo, F.M. Reliability and differentiation capability of dynamic and static kinematic measurements of rearfoot eversion in patellofemoral pain. Clin. Biomech. 2015,30, 144–148. [CrossRef] 33. Pawar, P.K.; Dadhich, A. Study of correlation between human height and hand length in residents of Mumbai. Int. J. Biol. Med. Res. 2012,3, 2072–2075. 34. Bujang, M.A.; Baharum, N. A simplified guide to determination of sample size requirements for estimating the value of intraclass correlation coefficient: A review. Arch. Orofac. Sci. 2017,12, 1–11. 35. Charan, J.; Biswas, T. How to calculate sample size for different study designs in medical research? Indian J. Psychol. Med. 2013,35, 121–126. [CrossRef] [PubMed] 36. Sample, R.B.; Jackson, K.; Kinney, A.L.; Diestelkamp, W.S.; Reinert, S.S.; Bigelow, K.E. Manual and cognitive dual tasks contribute to fall-risk differentiation in posturography measures. J. Appl. Biomech. 2016 , 32, 541–547. [CrossRef] [PubMed] 37. Scoppa, F.; Capra, R.; Gallamini, M.; Shiffer, R. Clinical stabilometry standardization: Basic definitions–acquisition interval–sampling frequency. Gait Posture 2013,37, 290–292. [CrossRef] 38. Huurnink, A.; Fransz, D.P.; Kingma, I.; van Dieën, J.H. Comparison of a laboratory grade force platform with a Nintendo Wii Balance Board on measurement of postural control in single-leg stance balance tasks. J. Biomech. 2013,46, 1392–1395. [CrossRef] 39. Duarte, M.; Freitas, S.M. Revision of posturography based on force plate for balance evaluation. Braz. J. Phys. Ther. 2010,14, 183–192. [CrossRef] 40. Bonnechere, B.; Jansen, B.; Jan, S.V.S. Cost-effective (gaming) motion and balance devices for functional assessment: Need or hype? J. Biomech. 2016,49, 2561–2565. [CrossRef] 41. Zhu, Y. Design and validation of a low-cost portable device to quantify postural stability. Sensors 2017 , 17, 619. [CrossRef] [PubMed] 42. Uebbing, T.J. User Experience in Smart Environments: Design and Prototyping. Master’s Thesis, University of Twente, Twente, The Netherlands, 2016. 43. Williams, Q.I.; Gunn, A.H.; Beaulieu, J.E.; Benas, B.C.; Buley, B.; Callahan, L.F.; Cantrell, J.; Genova, A.P.; Golightly, Y.M.; Goode, A.P.; et al. Physical therapy vs. internet-based exercise training (PATH-IN) for patients with knee osteoarthritis: Study protocol of a randomized controlled trial. BMC Musculoskelet. Disord. 2015 , 16, 264. [CrossRef] [PubMed] 44. Peydro de Moya, M.F.; Baydal Bertomeu, J.M.; Vivas Broseta, M.J. Evaluaci ó n y rehabilitaci ó n del equilibrio mediante posturografía. Rehabilitación2005,39, 315–323. [CrossRef] 45. Tesio, L.; Rota, V.; Longo, S.; Grzeda, M.T. Measuring standing balance in adults: Reliability and minimal real difference of 14 instrumental measures. Int. J. Rehabil. Res. 2013,36, 362–374. [CrossRef] 46. Hoving, J.L.; Pool, J.J.; van Mameren, H.; Devill é , W.J.; Assendelft, W.J.; de Vet, H.C.; de Winter, A.F.; Koes, B.W.; Bouter, L.M. Reproducibility of cervical range of motion in patients with neck pain. BMC Musculoskelet. Disord. 2005,6, 59. [CrossRef] Healthcare 2020,8, 402 17 of 18 47. Benvenuti, F.; Mecacci, R.; Gineprari, I.; Bandinelli, S.; Benvenuti, E.; Ferrucci, L.; Baroni, A.; Rabuffetti, M.; Hallett, M.; Dambrosia, J.M. Kinematic characteristics of standing disequilibrium: Reliability and validity of a posturographic protocol. Arch. Phys. Med. Rehabil. 1999,80, 278–287. [CrossRef] 48. Doyle, T.L.; Newton, R.U.; Burnett, A.F. Reliability of traditional and fractal dimension measures of quiet stance center of pressure in young, healthy people. Arch. Phys. Med. Rehabil. 2005 ,86, 2034–2040. [CrossRef] 49. Schuck, P.; Zwingmann, C. The ’smallest real difference’s a measure of sensitivity to change: A critical analysis. Int. J. Rehabil. Res. 2003,26, 85–91. 50. Tao, W.; Liu, T.; Zheng, R.; Feng, H. Gait analysis using wearable sensors. Sensors 2012 , 12, 2255–2283. [CrossRef] 51. Donoghue, D.; Stokes, E.K. How much change is true change? The minimum detectable change of the Berg Balance Scale in elderly people. J. Rehabil. Med. 2009,41, 343–346. [CrossRef] 52. Kovacs, F.M.; Abraira, V.; Royuela, A.; Corcoll, J.; Alegre, L.; Tom á s, M.; Mir, M.A.; Cano, A.; Muriel, A.; Zamora, J. Minimum detectable and minimal clinically important changes for pain in patients with nonspecific neck pain. BMC Musculoskelet. Disord. 2008,9, 43. [CrossRef] [PubMed] 53. Marchetti, G.F.; Lin, C.C.; Alghadir, A.; Whitney, S.L. Responsiveness and minimal detectable change of the dynamic gait index and functional gait index in persons with balance and vestibular disorders. J. Neurol. Phys. Ther. 2014,38, 119–124. [CrossRef] [PubMed] 54. Batterham, A.M.; Hopkins, W.G. Making meaningful inferences about magnitudes. Int. J. Sports Physiol. Perform. 2006,1, 50–57. [CrossRef] 55. Cicchetti, D.V. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol. Assess. 1994,6, 284. [CrossRef] 56. Hickey, S.; Ford, G.; Buckley, J.; O’connor, A.F. Unterberger stepping test: A useful indicator of peripheral vestibular dysfunction? J. Laryngol. Otol. 1990,104, 599–602. [CrossRef] 57. Bartual, J.; Pérez, N. El sistema vestibular y sus alteraciones. Tomo 1998,1, 21–22. 58. Shumway-Cook, A.; Brauer, S.; Woollacott, M. Predicting the probability for falls in community-dwelling older adults using the Timed Up & Go Test. Phys. Ther. 2000,80, 896–903. 59. Mart í nez Carrasco, Á . An á lisis del riesgo de ca í das en ancianos institucionalizados mediante escalas de marcha y equilibrio. Ph.D. Thesis, Universidad de Murcia, Murcia, Spain, 28 January 2016. 60. Vellas, B.J.; Wayne, S.J.; Romero, L.; Baumgartner, R.N.; Rubenstein, L.Z.; Garry, P.J. One-leg balance is an important predictor of injurious falls in older persons. J. Am. Geriatr. Soc. 1997,45, 735–738. [CrossRef] 61. Hansson, E.E.; Persson, L.; Malmström, E.M. Influence of vestibular rehabilitation on neck pain and cervical range of motion among patients with whiplash-associated disorder: A randomized controlled trial. J. Rehabil. Med. 2013,45, 906–910. [CrossRef] 62. Orejas, J.I.B.; Varea, J.A.; Rodrigo, J.V.; Navas, A.C. Resultados y seguimiento de la rehabilitaci ó n vestibular. Revista ORL 2020,11, 107–114. [CrossRef] 63. Hopkins, W.G. A spreadsheet for monitoring an individual’s changes and trend. Sportscience 2017,21, 10. 64. Hopkins, W.G. Rebranding MBI as magnitude-based decisions (MBD). Sportscience 2019,i-iii, 23. 65. Hopkins, W.G.; Marshall, S.W.; Batterham, A.M.; Hanin, J. Progressive statistics for studies in sports medicine and exercise science. Med. Sci. Sports Exerc. 2009,41, 3–13. [CrossRef] [PubMed] 66. Salavati, M.; Hadian, M.R.; Mazaheri, M.; Negahban, H.; Ebrahimi, I.; Talebian, S.; Jafari, A.H.; Sanjari, M.A.; Sohani, S.M.; Parnianpour, M. Test–retest reliabty of center of pressure measures of postural stability during quiet standing in a group with musculoskeletal disorders consisting of low back pain, anterior cruciate ligament injury and functional ankle instability. Gait Posture 2009,29, 460–464. [CrossRef] [PubMed] 67. Degani, A.M.; Leonard, C.T.; Danna-dos-Santos, A. The effects of early stages of aging on postural sway: A multiple domain balance assessment using a force platform. J. Biomech. 2017,64, 8–15. [CrossRef] 68. Hanes, D.A.; McCollum, G. Cognitive-vestibular interactions: A review of patient difficulties and possible mechanisms. J. Vestib. Res. 2006,16, 75–91. 69. Swanenburg, J.; de Bruin, E.D.; Stauffacher, M.; Mulder, T.; Uebelhart, D. Effects of exercise and nutrition on postural balance and risk of falling in elderly people with decreased bone mineral density: Randomized controlled trial pilot study. Clin. Rehabil. 2007,21, 523–534. [CrossRef] 70. Tsukamoto, H.F.; Costa, V.d.S.P.; Silva Junior, R.A.d.; Pelosi, G.; Marchiori, L.L.d.M.; Vaz, C.R.S.; Fernandes, K.B.P. Effectiveness of a vestibular rehabilitation protocol to improve the health-related quality of life and postural balance in patients with vertigo. Int. Arch. Otorhinolaryngol. 2015,19, 238–247. [CrossRef] Healthcare 2020,8, 402 18 of 18 71. Baloh, R.W.; Jacobson, K.M.; Enrietto, J.A.; Corona, S.; Honrubia, V. Balance disorders in older persons: Quantification with posturography. Otolaryngol. Head Neck Surg. 1998,119, 89–92. [CrossRef] 72. Kim, S.K.; Kim, Y.B.; Park, I.S.; Hong, S.J.; Kim, H.; Hong, S.M. Clinical analysis of dizzy patients with high levels of depression and anxiety. J. Audiol. Otol. 2016,20, 174–178. [CrossRef] 73. Yuan, Q.; Yu, L.; Shi, D.; Ke, X.; Zhang, H. Anxiety and depression among patients with different types of vestibular peripheral vertigo. Medicine 2015,94. [CrossRef] [PubMed] 74. Yardley, L.; Redfern, M.S. Psychological factors influencing recovery from balance disorders. J. Anxiety Disord. 2001,15, 107–119. [CrossRef] 75. De Vet, H.C.; Terwee, C.B. The minimal detectable change should not replace the minimal important difference. J. Clin. Epidemiol. 2010,63, 804. [CrossRef] [PubMed] 76. Hamburg, M.A.; Collins, F.S. The path to personalized medicine. N. Engl. J. Med. 2010 ,363, 301–304. [CrossRef] [PubMed] Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).