Psychometric properties, acceptability, and feasibility of the Stability, Agility, Strength (SAS) Mat for balance assessment with potential for training Authors: Georgia Fisher, Anna Stackpool, Tina Vickery, Alexander Bloomfield, Niclas Richter, Katharine Scrivener Published: 2026 ABSTRACT BackgroundExisting tests of tandem balance have several limitations, including limited sensitivity, floor and ceiling effects, and assistance-related bias. The Strength Agility Stability (SAS) Mat was developed to address these limitations and can be used as the SAS Measure of tandem balance, and a balance training tool. We assessed the Measure’s test–retest reliability and concurrent validity, and the Mat’s acceptability and feasibility as a balance assessment and training tool for healthy older adults.MethodsA convenience sample of 44 community-dwelling adults aged ≥ 65 years (mean age 72.59, SD 7.15), with independent mobility (± walking aid) and no severe medical conditions, completed two sessions 4–7 days apart. In session 1, participants completed the SAS Measure, the Four Stage Balance Test (FSBT), and Timed Up and Go Test (TUGT). In session 2, participants completed the SAS Measure and a survey about the SAS Mat as an assessment and potential training tool. Test-retest reliability and validity were measured with Intraclass and Spearman’s correlation coefficients. Descriptive statistics were calculated for survey responses on perceived acceptability and feasibility of the SAS Mat.ResultsThe SAS Measure showed moderate test-retest reliability (Spearman’s r = 0.71, p < 0.001, 95%CI: 0.51 to 0.83; ICC[3,1] = 0.65, p < 0.001, 95%CI: 0.43 to 0.80). Concurrent validity was good with the FSBT (r = 0.67, p < 0.001, 95%CI: 0.46 to 0.81) and acceptable with the TUGT (r=-0.42, p = 0.014, 95%CI: − 0.64 to − 0.14). Most participants rated the SAS Measure as “completely acceptable” (90%, 37/41), believed the SAS Mat would improve balance (79%, 33/42), and expressed confidence in home use for balance training (88%, 37/42).ConclusionPreliminary evidence supports the validity and reliability of the SAS Measure as an emerging balance assessment tool, with promising acceptability and feasibility for assessment and potential training use. Future research should investigate its effectiveness as a balance training tool.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12877-026-07757-4. Background Maintaining good balance is key to the preservation of functional independence and quality of life in aging populations, alongside the prevention of falls [1]. Tandem stance is an important component of static balance and is thus essential to accurately assess and train in older adults. Postural sway is the gold-standard measure of tandem balance, where displacement of the Center of Pressure (CoP) is quantified by stabilometry [2]. However, this technology is rarely available in clinical settings. As such, a simpler, and more commonly used measure of tandem balance is the Four Stage Balance Test (FSBT) [3]. Here, a person moves through four stances of increasing difficulty: (1) parallel stance; (2) semi-tandem stance; (3) tandem stance; and (4) single-leg stance (SLS). They are scored from one to four, indicating the most difficult stance that they could hold independently for 10 s. A score of less than three has been associated with falls [4–6] and the need for a walking aid [7]. A score of less than four is associated with a higher risk of mortality [8]. Despite its widespread use, the FSBT is not without issue. First, its short ordinal scale brings inherent limitations: while each score is equidistant, the increase in difficulty between levels may not be uniform [9] and the scoring structure reduces sensitivity to change and increases potential floor or ceiling effects [10]. Second, some clinical FSBT instructions [11, 12] allow assistance to achieve each stance, and only then attempt to hold it independently. This assistance is consequential; the balance of people requiring support to achieve tandem stance is not statistically different to those unable to perform a fully supported tandem stance [13]. There is no place to record level of assistance in typical FSBT scoresheets (e.g [12]). ; as such, scores may be inflated or confounded. Third, the test-retest reliability of the FSBT in healthy older adults (r = 0.66) has only been assessed over sessions spaced six-months apart and is thus sub-optimal to guide clinical use [3], particularly where clinicians seek to evaluate the longitudinal impact of balance training. In response, the Strength Agility Stability (SAS) Mat was developed, a physiotherapist-designed, rubber backed mat with a suede non-slip top [14]. The upper side of the Mat contains markings that enable clinicians to assess tandem balance (the SAS Measure), and can also act a training tool, assisting the independent performance of balance exercises. During the SAS Measure, a person steps through similar positions to the FSBT, standardised by footprint-shaped markings on the mat. The scoring criteria contain strict instructions about support allowed at each stage and scores on a continuous scale from 0 to 70, which incorporates the time spent in a particular stance. When used for training, the Mat can be used to guide foot position for a variety of balance exercises [15]. As a portable and relatively low-cost tool, the SAS Mat may offer potential for home-based balance assessment and training. Newly developed, the SAS Measure has yet to undergo psychometric testing to establish validity and reliability. In addition, the feasibility and acceptability of the SAS Measure and SAS Mat as balance assessment and training tools, from the perspective of older people who will use them, are not yet established. Thus, we aimed to conduct a preliminary evaluation of the psychometric properties of the SAS Measure and the acceptability and feasibility of the SAS Mat as a balance assessment and potential training tool. Our primary aim was to assess the test-retest reliability and validity of the SAS Measure as an assessment of balance for healthy aged individuals. Our secondary aim was to describe the perceptions of healthy aged individuals about the acceptability and feasibility of the SAS Measure as an assessment of balance, and the SAS Mat as a tool for home balance training. Methods Design A multi-method study was undertaken in collaboration with the SAS Mat creator (AS). Ethical approval was granted by the Macquarie University Human Research Ethics Committee (ref no. 520251857965351) and all participants provided informed written consent. The study is reported according to the STROBE checklist for cohort studies (Additional File 1) [16]. Participants A convenience sample was recruited between February and October 2025 via advertisement to community groups, medical clinics, and volunteer groups. Participants comprised independently mobile, community-dwelling adults aged \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\ge\:$$\end{document}65 with no current, severe medical conditions. People with major lower limb or spinal injury or surgery in the past 12 months, a confirmed medical diagnosis of severe cognitive impairment or a major neurological disorder, or previous experience using the SAS Mat were excluded from the study. All participants completed an online screening form; eligibility was confirmed via telephone or email. Study assessments Timed Up and Go Test (TUGT) The TUGT is a validated, and commonly used assessment of balance and functional mobility for community dwelling older adults [17]. Briefly, the test commenced with a participant sitting in a standard height chair with their back against the chair back. When instructed to ‘Go’ the participant stood from the chair, walked three meters, turned, walked back to the chair and sat down. Timing (in seconds) began at the instruction of ‘Go’ and stopped when the participant returned to a seated position. A practice was provided prior to the actual test. Four Stage Balance Test (FSBT) In the FSBT, participants moved through four stances of increasing difficulty: Level (1) feet together (parallel stance); Level (2) semi-tandem stance (instep of one foot touching the big toe of the other); Level (3) tandem stance (heel-to-toe); and Level (4) SLS (self-selected) [3]. The participant was required to hold each stance for 10s without foot movement or support before progressing to the next. Scores ranged from one to four, indicating the highest level able to be maintained independently for 10s. Participants were not provided with assistance to attain any stance. SAS Measure During the SAS Measure, a participant progressed through a maximum of six test levels. The SAS Mat (Fig. 1) was placed on the floor with one long side 40 cm from a wall and a backward-facing chair 40 cm from the opposite side. The participant started with one foot (self-selected) on the starting footprint and the other foot shoulder-width apart. The therapist then instructed the participant “Without touching the wall or chair, step your other foot onto the [colour] footprint”. For Level 1, the participant could be assisted into the test position if needed. If the participant slid or shuffled their foot into position, rather than taking a step, the physiotherapist asked them to repeat the level. If the participant could hold a position for 10 s without assistance, they returned to the starting position and moved to the next level. After Level 1, assistance to achieve any test position was not permitted. The test ceased when the participant could either not achieve or hold a test position for 10 s. The SAS Mat is scored on a continuous 0–70 range to capture both the highest level achieved and the time spent in each position, thereby potentially improving sensitivity to subtle differences in balance performance compared with ordinal scoring systems that record only the final level attained. Figure 1 describes the procedure and scoring structure of the SAS Mat in more detail. Fig. 1Image adapted from [15]. A: The SAS Mat. The dotted black line denotes the Assessment Zone and is not visible on the physical Mat. B: SAS Measure positions. Levels 1 to Level 4: Grey foot = grounded foot position. Coloured foot = moving foot end position. The participant started with a self-selected grounded foot in position. Then, they stepped unassisted with the moving foot onto the coloured footprint and attempted to hold for 10 s. Time in standing with no support was recorded. If 10 s was achieved, the participant moved to the next level. NB If the participant required assistance to successfully step into the Level 1 position but could hold the position for 10 s, they were scored as a Level 1 A and did not proceed. Level 5 and Level 6: Coloured foot = test foot. At each level, the participant started with a self-selected foot on the purple footprint. They then lifted the other foot to achieve a single leg stance (SLS) and attempted to hold for 10 s. Next, they repeated this process, with the other foot becoming the test foot. Time in unassisted SLS was recorded for both feet. C: SAS Measure scoring criteria. t = time spent in position in seconds, always ranged from 1 to 10 s. Lt = time spent in single leg stance on left leg. Rt = time spent in single leg stance on right leg. SLS = single leg stance, s = seconds, support = upper-limb contact with wall or chair, or assistance from physiotherapist Image adapted from [15]. A: The SAS Mat. The dotted black line denotes the Assessment Zone and is not visible on the physical Mat. B: SAS Measure positions. Levels 1 to Level 4: Grey foot = grounded foot position. Coloured foot = moving foot end position. The participant started with a self-selected grounded foot in position. Then, they stepped unassisted with the moving foot onto the coloured footprint and attempted to hold for 10 s. Time in standing with no support was recorded. If 10 s was achieved, the participant moved to the next level. NB If the participant required assistance to successfully step into the Level 1 position but could hold the position for 10 s, they were scored as a Level 1 A and did not proceed. Level 5 and Level 6: Coloured foot = test foot. At each level, the participant started with a self-selected foot on the purple footprint. They then lifted the other foot to achieve a single leg stance (SLS) and attempted to hold for 10 s. Next, they repeated this process, with the other foot becoming the test foot. Time in unassisted SLS was recorded for both feet. C: SAS Measure scoring criteria. t = time spent in position in seconds, always ranged from 1 to 10 s. Lt = time spent in single leg stance on left leg. Rt = time spent in single leg stance on right leg. SLS = single leg stance, s = seconds, support = upper-limb contact with wall or chair, or assistance from physiotherapist Survey The online survey (REDCap) was purpose-built and contained questions based on the Theoretical Framework of Acceptability [18]. Comprising 15 five-point Likert scale questions and one numeric slider scale, asking participant impressions of the SAS Mat as an assessment and training tool. The survey contained embedded videos about how to train with the SAS Mat. Participants watched the videos before completing questions about training with the Mat. Participants were informed that the mat costs approximately $189 AUD and were asked whether they would purchase it at that price. The survey was designed and then piloted by the study research and clinical team. A copy of the survey is available in Additional File 2, and the videos on SAS Mat training can be viewed in Additional File 3. Procedures Each participant completed two assessment sessions (session 1 and session 2), four to seven days apart. All sessions were conducted in the same clinic room at MQ Hospital outpatient physiotherapy department by one of two senior physiotherapists (AB, NR). The study physiotherapists received one-hour of training on the SAS Mat assessment, conducted by AS. Study physiotherapist competency in the SAS Mat assessment was signed off on by AS before the commencement of data collection, and they were able to request additional clarification on assessment conduct prior to the first study session. For each participant, both sessions were conducted by the same physiotherapist. In session 1, participants provided basic demographic information and then completed the TUGT, FSBT, and the SAS Measure. The sequence of testing was randomly allocated to remove any consistent effect of fatigue, warm-up, or test order. Study physiotherapists were not blinded to previous scores, as the same physiotherapist conducted both sessions for a given participant. Study physiotherapists were unable to access session 1 data after entry; any data entry issues were rectified by a member of the research team (TV). In session 2, participants repeated the SAS Measure. Participants then completed the study survey. Statistical analysis and sample size To assess test-retest reliability, we calculated a correlation coefficient and intraclass correlation coefficients (ICC [1, 3] and ICC[3,k]) of SAS Measure scores in session 1 and session 2. Prior to ICC analysis, SAS Measure scores were inspected for outliers and approximate normality. To assess validity, we calculated the correlation coefficients between the session 1 SAS Measure score and the Four Stage Balance Test and TUGT scores respectively. We also calculated the correlation between TUGT and FSBT scores. As data were non-normal or ordinal, Spearman’s correlations were calculated. Participants with missing data points for a particular outcome measure were removed from the associated analyses. Given that the SAS Measure was new to the study physiotherapists, we also ran independent sample t-tests comparing SAS Measure scores for the first 20 participants and the final 20 participants, to investigate any consistent scoring differences over time. Finally, we calculated descriptive statistics of participant demographics and survey results. Analyses were conducted using custom built Python scripts (see end of article). Bonferroni corrections were applied for multiple correlations. The data analyst (GF) was blind to all participant results until analysis, and analyses were checked using Excel by a second researcher (TV). Assuming a predicted ICC of > 0.9 (a conservative estimate given the Berg Balance Scale has test-retest reliability of 0.98 [19]), 80% power, and confidence interval precision of 0.15, the required sample for test-retest reliability was n = 40 participants [20]. Assuming a conservative estimate of a minimum predicted correlation coefficient of 0.45, and a desired power of 80% and alpha of 0.05, the required sample size for validity was 36 participants [21, 22]. We thus aimed to recruit a sample of n = 45 participants to account for potential participant dropout / non-completion. Results A total of 45 participants were recruited and completed the study. One participant was excluded from analyses due to a conflict with inclusion criteria discovered post-data collection. Two participants did not complete study session 2 due to physiotherapist illness. One participant’s session 2 SAS Measure score was removed, and another’s TUGT score was removed due to data collection issues. One participant did not complete the assessment component of the survey, and five questions contained missing data from either one or two participants. Participant demographics are reported in Table 1. Table 1Demographic data. # Counts do not add to 44 as participants could declare more than one health conditionVariableResult N Age (years) Mean 72.5944 SD 7.15Height (cm) Median 162.043 IQR 157.0 to 171.0 Weight (kg) Median 67.043 IQR 61.0 to 78.0BMI Median 24.6543 IQR 23.52 to 26.54N medicines Median 2.044 IQR 1.0 to 3.25 Female n = 34 43 Male n = 9 Living independently n = 39 44 Some care provided n = 5Fall in last year Yes n = 10 44 No n = 34N falls in last year One n = 7 10 Two or more n = 3 Walks independently n = 43 44 Uses walking stick n = 1 Musculoskeletal problems n = 32 44# Eye/vision problems n = 27 Chronic illness n = 17 Heart disease n = 4 Chronic dizziness n = 3 Diabetes n = 2 Other n = 11 None n = 2 Table 1Demographic data. # Counts do not add to 44 as participants could declare more than one health conditionVariableResult N Age (years) Mean 72.5944 SD 7.15Height (cm) Median 162.043 IQR 157.0 to 171.0 Weight (kg) Median 67.043 IQR 61.0 to 78.0BMI Median 24.6543 IQR 23.52 to 26.54N medicines Median 2.044 IQR 1.0 to 3.25 Female n = 34 43 Male n = 9 Living independently n = 39 44 Some care provided n = 5Fall in last year Yes n = 10 44 No n = 34N falls in last year One n = 7 10 Two or more n = 3 Walks independently n = 43 44 Uses walking stick n = 1 Musculoskeletal problems n = 32 44# Eye/vision problems n = 27 Chronic illness n = 17 Heart disease n = 4 Chronic dizziness n = 3 Diabetes n = 2 Other n = 11 None n = 2 Demographic data. # Counts do not add to 44 as participants could declare more than one health condition Reliability of the SAS Measure There was no statistical difference in the SAS Measure scoring of the first and the last 20 participants (session 1: t = 0.29, p = 0.77; session 2: t = -1.83, p = 0.08). The SAS Measure showed moderate single-measure test-retest reliability (n = 41; ICC [1, 3] = 0.65, p < 0.001, 95%CI: 0.43 to 0.80; Spearman’s r = 0.71, p < 0.001, 95%CI: 0.51 to 0.83). When considering the average of the session 1 and session 2 SAS Measure scores, reliability was good (n = 42; ICC[3,k] = 0.79, p < 0.001, 95%CI: 0.60 to 0.89). Validity of the SAS Measure Concurrent validity of the SAS Measure was good with the FSBT (n = 44; Spearman’s r = 0.67, p < 0.001, 95% CI: 0.46 to 0.81). Concurrent validity of TUGT was acceptable with the SAS Measure (n = 43; Spearman’s r = − 0.42, p = 0.014, 95% CI: − 0.64 to − 0.14) and the FSBT (n = 43; Spearman’s r = − 0.39, p = 0.02, 95%CI: − 0.62 to − 0.10). Figure 2 shows plots of data used for all correlations. Fig. 2Plots of data used for psychometric testing of the SAS Measure. A: Session 1 SAS Measure score vs. session 2 SAS Measure score (n = 41); B: Session 1 SAS Measure Score vs. TUGT score (n = 43); C: Session 1 SAS Measure score vs. FSBT Score (n = 44); D: TUGT Score vs. FSBT Score (n = 43). Grey dots represent individual participant data points. Blue lines represent the Ordinary Least Squared regressions for plots of two continuous outcomes. Note that FSBT scores were ordinal and ranged from 1 to 4, but that individual scores have been vertically scattered in panels C and D to show all data points. Two participants (4.55%) participants achieved the minimum FSBT score of 1, 26 participants (59.09%) achieved the maximum FSBT score of 4. No participants achieved the minimum or maximum SAS Mat Score in Session 1 Plots of data used for psychometric testing of the SAS Measure. A: Session 1 SAS Measure score vs. session 2 SAS Measure score (n = 41); B: Session 1 SAS Measure Score vs. TUGT score (n = 43); C: Session 1 SAS Measure score vs. FSBT Score (n = 44); D: TUGT Score vs. FSBT Score (n = 43). Grey dots represent individual participant data points. Blue lines represent the Ordinary Least Squared regressions for plots of two continuous outcomes. Note that FSBT scores were ordinal and ranged from 1 to 4, but that individual scores have been vertically scattered in panels C and D to show all data points. Two participants (4.55%) participants achieved the minimum FSBT score of 1, 26 participants (59.09%) achieved the maximum FSBT score of 4. No participants achieved the minimum or maximum SAS Mat Score in Session 1 Survey Participants reported feeling comfortable (n = 41, 100%) and confident (n = 40, 98%) having their balance assessed with the SAS Measure and agreed that it did a good job in assessing their balance (n = 34, 85%). Participants reported feeling that the assessment took a little effort (n = 28, 68%) or no effort at all (n = 11, 27%) to complete, and that the tool was a fair (n = 32, 78%) and acceptable (n = 40, 98%) measure of balance for healthy older adults. Participants agreed that the SAS Mat was a fair (n = 36, 85%) and acceptable (n = 37, 88%) way of training balance at home. Participants reported that they would feel confident (n = 37, 88%) and comfortable (n = 39, 93%) to use the Mat independently at home. Most participants felt that it was clear how training with the Mat would help improve their balance (n = 33, 83%) and that home-use of the Mat would not interfere with their other priorities (n = 29, 79%). The mean reported likelihood of home use of the SAS Mat for balance training was 72% (SD 29%), however, most participants indicated that they were not willing to pay $180 for a Mat (n = 29, 73%). Further details are shown in Additional File 4 and Additional File 5. Discussion Tandem stance is an important component of any balance assessment; however, common clinical tests of tandem balance have limitations. The SAS Mat is a novel assessment designed to overcome these limitations when measuring individuals’ ability to tandem stand, with a dual function as a balance training tool. In this study, we conducted a preliminary investigation of the psychometric properties of the SAS Measure in a sample of healthy, community-dwelling older people. We also investigated the feasibility and acceptability of the SAS Mat as an assessment and potential training tool. The SAS Measure showed moderate to good test–retest reliability and acceptable concurrent validity with established clinical measures assessing related balance constructs. Most participants found the SAS Measure to be an acceptable assessment of balance and felt confident during assessment. A majority believed home-based SAS Mat training would improve their balance, would feel confident and comfortable to use the Mat independently at home. In contrast to otherwise positive usability and acceptability ratings, the cost of the mat was prohibitive, with most participants reporting an unwillingness to pay the stated price of the mat. This may represent a substantial barrier to wide-spread adoption of the mat. When used as an assessment of balance at a single time point, the SAS Measure demonstrated moderate test–retest reliability, consistent with interpretive thresholds in which ICC values between 0.50 and 0.75 indicate moderate reliability. Although an ICC > 0.9 was assumed for sample size estimation based on established balance measures, this proved optimistic for a novel tool with different constructs and task demands. The observed moderate single-measure reliability (ICC [1, 3] = 0.65) reflects early-stage psychometric evaluation but improved to a good level when scores from two sessions were averaged (ICC[3,k] = 0.79), supporting use in research contexts or where repeated measurement is feasible. Accordingly, the SAS Measure is not recommended as a standalone indicator of balance ability but may be used as a screening measure and combined with other assessments to provide a more holistic evaluation of balance impairment, as commonly recommended for balance testing [23–25]. While Bonferroni corrections were applied to reduce the risk of Type I error due to multiple correlations, this conservative approach may have reduced statistical power and should be considered when interpreting the magnitude and significance of the observed associations. The SAS Measure could be used in the place of the FSBT in test batteries, offering an alternative to the limitations of the FSBT ordinal scoring system. Additionally, when the average of two assessments was instead used, the SAS Measure’s test-retest reliability improved. Thus, clinicians could consider using this strategy to more reliably assess tandem balance ability. Notably, current instructions for the SAS Measure do not standardise practice before formal testing. Due to the influence of confidence and learning-effects on test performance, many clinical balance assessments recommend the use of a practice test, or the best score from a series of attempts [26–28]. These strategies could be used to improve the test-retest reliability of the SAS Measure as a point-in-time assessment, a possibility that should be further investigated. Validity assessment in this study was limited to correlations with commonly used clinical measures of balance and functional mobility. While this approach is appropriate for preliminary evaluation, neither the FSBT nor the TUGT represents a gold-standard measure of tandem balance, and both assess related but distinct constructs. The absence of an objective reference standard, such as force plate–based stabilometry, means that criterion validity could not be evaluated and should be addressed in future research. In many balance tests, a hold time of 10 s is specified to complete a test position [3, 29]; however, this duration may be insufficient to identify subtle balance deficits in community-dwelling older adults [30]. In a longitudinal study, de Abreu and colleagues found that older adults could hold tandem stance for ~ 22 s and single-leg stance for ~ 15 s, and that each additional second of hold time was associated with a 5% reduction in fall risk over six months [30]. Thus, while lower FSBT scores are associated with falls [6] and walking aid use [7], the test may fail to detect more subtle impairments. Rather than extending hold times alone, the SAS Measure increases sensitivity by scoring single-leg stance on both limbs and under vision-removed conditions, targeting balance during movement and position change rather than quiet standing. Consistent with this, 26 participants achieved the maximum FSBT score in our sample, whereas none achieved the maximum SAS Measure score, indicating an absence of ceiling effects. Among those with maximum FSBT scores, 35% could not hold single-leg stance with vision on both limbs, and 65% could not do so without vision. Together, these findings suggest that the SAS Measure may offer improved sensitivity to early balance impairment than ordinal tandem balance tests, supporting its potential utility for screening and monitoring balance in high-functioning older adults in community and outpatient settings. Limitations of this study include the use of a convenience sample of older adults, which may have introduced selection bias and limited the representativeness of the sample. Participants were likely to be relatively healthy, motivated, and functionally independent, which may have contributed to the high overall performance observed and restricts the generalisability of the findings. As such, the results are most applicable to healthy, community-dwelling older adults with independent mobility and cannot be extrapolated to clinical or more heterogeneous populations, such as frail older adults or those with mobility impairments. In this high-functioning sample, performance on the FSBT demonstrated a ceiling effect, constraining interpretation of the correlation analyses. In addition, study physiotherapists were not blinded to previous testing sessions, which may have influenced repeat SAS Measure scoring and further contributed to inflated test–retest reliability. The retest interval of four to seven days may also have contributed to learning or recall effects, potentially inflating reliability estimates; longer intervals should be considered in future studies. Additionally, acceptability ratings were collected after participants viewed a training video, which may not fully reflect real-world use. Future research should therefore assess acceptability following independent, unsupervised home implementation and evaluate performance in more diverse and clinically representative populations. Conclusion Preliminary evidence supports the concurrent validity and test–retest reliability of the SAS Measure as an emerging balance assessment tool in healthy older adults. The SAS Mat also showed promising participant acceptability and feasibility for assessment and potential training use. Future research should investigate its effectiveness as a balance training intervention, and its assessment properties in other clinical populations. Supplementary Information Additional File 1. Additional File 1. Additional File 2. Additional File 2. Additional File 3. Additional File 3. Additional File 4. Additional File 4. Additional File 5. Additional File 5.