This study aimed to determine the test-retest reliability of the BESTest in youth with IDs. Overall, the BESTest appeared to be a reliable test for assessing postural control in youth with IDs for epidemiological research aimed at determining the effects of interventions designed to improve physical function in this population. To the authors’ knowledge, no studies have considered the test-retest reliability of the BESTest in youth with IDs. Postural control tests that have low or otherwise senseless reliability coefficients should not be used to assess postural control in people with IDs. Given that existing tools for assessing clinical balance cannot assist the therapist in diagnosing the underlying causes of balance disorder, BESTest can distinguish balance subsystems and can purposefully plan the treatment process.21.
Several statistical methods and indices have been proposed to test the reliability of outcome assessment, but it was forcefully suggested that both relative and absolute reliability should be introduced 23. For relative reliability, the ICC is the most generally used statistical method because it shows the level of agreement among tests. For absolute reliability, the SEM can reflect the reliability of the response with a 95% CI, which provides useful information relevant to the actual value of real change24.
Six subsystems underlying control of balance that are targeted in the new Balance Evaluation Systems Test (BESTest) involve the integration of biomechanical constraints, stability limits and verticality, anticipatory postural adjustments, postural responses, sensory orientation, and stability in gait regarding body position and motion with the ability to generate forces to control body position. Each system consists of neurophysiological mechanisms that control a particular aspect of postural control. Our results are consistent with findings from studies in other various populations, such as people with stroke, Parkinson’s disease, spinal cord injury, and multiple sclerosis, and also in people with balance disorders and people with an increased risk of falling 25–30. These studies provide ICC values for total scores ranging from.80 to.97, measurement errors ranging from and MDC ranging from 2.4 to 5.2.
In our study, excellent reliability was found for both the biomechanical constraints, transitions and anticipatory, sensory orientation, and stability in gait sections, which is consistent with the results from studies in people with Parkinson’s disease 27 and Multiple Sclerosis30, and in people with increased risk of falling.
On the other hand, studies such as Villamonte et al. measured test-retest reliability scores on 16 balance tests in children, teenagers, and young adults with ID. Among the tests conducted are the STS (20 s long) and the TUG (9 m distance), the former being reliable (ICC > 50) only in young women and young men, while the latter is not reliable for any group of ID. Among the limitations of the study are its small sample (21 people) and very varied age range (5–31 years old)31.
PH Boer and S.J. Moss explored the test-retest reliability and minimal detectable change of selected functional fitness test items in adults with ID. The results indicated all tests showed excellent results (ICCs > 0.9). All SEM values demonstrated acceptable measurement precision (SEM < SD/2). Values for MDC90 are provided for all 12 tests 32.
The Consensus-Based Standards for the Selection of Health Measurement Instruments initiative defines domain reliability as "the degree to which the measurement is free from measurement error," which implies that scores for unchanged patients are constant in repeated measurements. Estimations of the MDC and limits of agreement (Fig. 1) are important to define the change and the boundaries that need to be exceeded to show a change beyond the measurement error, that is, a "true" change. The current study's findings indicate that a change of 10 points or more in the BESTest total score is required to determine a "true" change in balance control in IDs. However, if the sample consists of only IDs, a change of 10 points or more is needed.
In this study, the kappa values for all items except items 2 and 4 were considered excellent. The kappa value for items 2 and 4 included in the section on stability limits/ verticality, and reactive postural control was fair to good. This contributed to the lower reliability found in this section compared with the other BESTest sections. The assessment of items 2 and 4 on the first test occasion seemingly caused a learning effect that biased the assessment on the second test occasion; that is, the learning effect caused anticipation when the item was reassessed, which made the performance of the item less reactive and more proactive 30.
The results of this investigation suggest that the BESTest is a reliable test that can be used to assess overall postural control in youth with IDs. This study has several strengths, including the use of a single researcher to collect all of the data to ensure the consistency of the assessments. Nevertheless, there are some limitations to this study. First, results from this study are valid for individuals with IDs; they cannot be generalized to subjects with other sub-types of IDs. Second, data were not collected on socioeconomic status or sports that the participants practiced. These factors may influence postural control levels. The average age of the participants was 21.1 years old, which may have influenced postural control as older youths may have performed better on the BESTest when compared to adults and individuals. Due to the potential impact that balance proficiency has on physical activity levels and the overall health of individuals, including those with IDs, it is paramount for clinicians and researchers to determine if these tests are reliable. The BESTest, which is often used in a variety of settings, might be a feasible assessment tool.