Postural Control Measures in Randomized Controlled Trials for Older Adults Balance: A Systematic Scoping Review
Article information
Abstract
The aim of this work was to identify and characterize the measures employed for assessing postural control in randomized controlled trials (RCTs) of balance interventions in older adults with the reference to the Systems Framework for Postural Control. A scoping review was conducted, and RCTs of balance interventions in older adults published from 2013 to March 2023 were considered for inclusion. Two hundred and seventy-one studies were included with a total of 49 different measures used; the Timed Up and Go test being the most commonly employed. The median number of components of postural control assessed per study was five. The most frequently assessed components were motor systems and static stability, while reactive postural control, cognitive influences and verticality were the least frequently assessed. Postural control in RCTs of balance in older adults was assessed using a wide range of measures, but also from the perspective of a limited number of components.
INTRODUCTION
Postural control, which can be defined as the act of maintaining, achieving or restoring a state of balance during any posture or activity, is a major concern in older adults.1,2) Age-related changes—such as declines in muscular strength, sensory function, and cognitive processing—can negatively affect balance and increase the risk of falls.3) Moreover, the presence of comorbidities frequently encountered in later life may further impair postural control.4,5) Yet, postural control is a complex concept, requiring efficient interactions between several systems of the human body.6,7) The accurate evaluation of postural control is a fundamental basis for implementing balance training programs.8) The high complexity of the postural control system requires a comprehensive assessment to be relevant. Horak7) has highlighted the need for comprehensive standardized measures that assess the different components of postural control. Therefore, this author has proposed the “Systems Framework for Postural Control” as a conceptual basis for this purpose.7) Later modified by Sibley et al.9) this conceptual framework considers nine components of postural control—Static Stability, Underlying Motor Systems, Functional Stability Limits, Verticality, Reactive Postural Control, Anticipatory Postural Control, Dynamic Stability, Sensory Integration, and Cognitive Influences. The detailed description of the conceptual framework and its components is available in Supplementary Table S1.
Over the past decade, a large number of studies have focused on interventions designed to improve postural control in older adults, leading to systematic reviews and practical recommendations.10,11) However, particular attention needs to be paid to how postural control was assessed in these studies. This includes not only the psychometric qualities of the assessments used but also the conceptual approach. Previous scoping reviews have characterized the components of postural control in standardized balance assessment measures for the pediatric population,12) people with chronic obstructive pulmonary disease13) and people with stroke.14) However, to our knowledge, no work has summarized and characterized components of postural control in assessment measures used in interventional trials carried out with older adults. This would enable us to better understand their alignment with the established conceptual knowledge on postural control and to identify gaps in the literature regarding this large and heterogeneous population. In order to achieve this, a scoping review represents an appropriate methodology for a literature review.15) Moreover, this study will consider the measures used to assess postural control in randomized controlled trials (RCTs), which are widely regarded as the gold standard design for assessing the efficacy of interventions. Therefore, the aim of this scoping review was to identify and characterize the standardized measures employed for assessing postural control in RCTs of interventions aimed at improving postural control in older people over the last decade with the reference to the Systems Framework for Postural Control.9)
The review’s questions were:
What were the standardized measures employed for assessing postural control in older adults?
Which components of postural control were assessed by these standardized measures?
MATERIAL AND METHODS
A systematic scoping review was carried out in accordance with the recommendations of the Johanna Briggs Institute (JBI).16) This review was reported according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA - ScR) (corresponding checklist is available in Supplementary Table S2). The review protocol was a priori registered on Open Science Framework, under the number JV7K9 (doi.org/10.17605/OSF.IO/JV7K9).
Inclusion Criteria
Participants
Studies including participants aged 60 years and older, regardless of their health status, were eligible.
Concept
This review explored the concept of postural control and its components, as proposed by Horak's systems framework for postural control7) and later adapted by Sibley et al.9) to establish an operational definition of the mode. The aim of this review was to identify and list the standardized measures used to assess postural control. To be considered as “standardized,” a measure should have both a standardized testing protocol and a standardized evaluation criteria.9)
Context
Postural control is a significant concern in this population,17) and research on interventions to improve it remains relevant today as there is no clear consensus on a gold-standard intervention. A thorough assessment of postural control is crucial to the implementation of balance training programs. The aim was to assess whether the information available from interventional studies captured all the conceptual dimensions involved in this complex activity.
Types of evidence sources
This scoping review was limited to randomized controlled trials that were specifically designed to evaluate interventions for improving postural control.
Search Strategy
The search strategy was not restricted by geography and covered the period from 2013 to 2023, providing an overview of the last decade of research in the field. The search strategy followed the three steps recommended by the JBI16):
Step 1: An initial search was conducted on Medline and Cochrane Library to identify articles related to our research question. The search was performed using the terms “postural balance” and “elderly people.” The text words in titles and abstracts, as well as different keywords, were used to develop a complete search strategy in the second step.
Step 2: The search strategy for postural control components included information from the first step and specific keywords. The search was conducted on various databases, including Medline, Google Scholar, SciELO, Cochrane Library, PEDro, and clinicaltrial.gov. The strategy was adapted as necessary for each database and is presented in Supplementary Table S3. For the Google Scholar search, the first 500 results were screened.18)
Step 3: A search was conducted for additional sources in the references list of identified articles. If necessary, the reviewers contacted the authors for further information. For conference abstracts and grey literature, a systematic attempt was made to identify whether a peer-reviewed full text exists or if the submission process is still ongoing, and whether the study was appropriately registered.19) Studies published in English and French from 2013 and up to March 6, 2023 were considered for inclusion as the reviewer team was not qualified in other languages.
Source of Evidence Selection
After conducting the search, all identified citations were uploaded into the Rayyan application. The review team (consisting of R.P., B.L., and K.J.) independently reviewed the titles, abstracts, and keywords of the identified citations and applied the inclusion and exclusion criteria. In cases where information was incomplete, the full text was studied. The review team then independently examined the full texts, applying the inclusion and exclusion criteria. In cases of disagreement, an objective discussion was held between the parties. Reasons for excluding sources that did not meet the inclusion criteria were documented and presented in the final review. The full search results were reported in the final report and presented in a flow diagram following the PRISMA-ScR guidelines.20)
Data Extraction
The scoping review extracted relevant data using a charting table that was developed in connection with the review questions and objectives. The table included information on the studies (country, authors, year of publication), population (number, age, gender of participants), and postural control outcomes (primary and secondary outcomes). For each standardized measure of postural control, the detail of postural control components covered was established using the previous work of Sibley et al.9) and completed by review team during specific meetings if necessary. The charting table underwent pilot testing by three reviewers who randomly selected three sources and was modified during the review process to ensure the most suitable data extraction. The charting table is presented in Supplementary Table S4.
Analysis of the Data
Descriptive analysis of included studies was performed for the year of publication, country origin, participants, postural control standardized measures and postural control components. A narrative synthesis, tables and figures of the results from the included studies were realized for studies characteristics and key findings.
RESULTS
Study Inclusion
A total of 271 studies were included in this scoping review, the details of which are illustrated in Fig. 1.
Characteristics of Studies
Among the 271 included RCTs (the complete list of sources and their detailed characteristics is provided in Supplementary Table S5) that focused on postural control in the older adults, the median sample size was 50 (34–73) (Table 1). Approximately one in six studies (n=47, 17.3%) focused exclusively on women, while others (n=5, 1.8%) focused exclusively on men. The weighted median mean age of these samples was 73.5 years (interquartile range [IQR], 70–78.9). None of the studies gave special attention to the “oldest old” population, although 15.2% (n=41) had a mean age above 80 years old. Seventy-six percent of RCTs (n=41, 15.2%) were conducted on the general population. Among the studies that selected participants based on specific conditions, the three most frequent clinical settings were related to osteoporosis/fractures (n=14, 5.2%), geriatric syndromes (n=11, 4.1%), and cognition/dementia (n=10, 3.7%). The included studies were conducted in all continents except Africa. The most represented countries were Brazil (n=29), USA (n=27), Spain (n=23), Iran and South Korea (n=21). The full distribution of countries of included studies is available in Supplementary Table S6.
Key Findings
Postural control standardized measures
In this review, a total of 49 standardized measures of postural control were identified in the included studies (the full list can be consulted in Supplementary Table S7) with a median number of 2 measures used per study (IQR, 1–3). The Timed Up and Go test (TUG) was the most frequently used, evaluated in 46.1% of included studies (n=125), followed by the Berg Balance Scale (BBS) (n=75, 27.7%), the static force platform eyes-open test (n=67, 24.7%), and the single-leg test (n=51, 18.9%) (Fig. 2). Detailed components of postural control covered by each examined measure is provided in Supplementary Table S8.
Components of postural control
Fig. 3 provides an overview of the postural control components that were assessed in the 271 RCTs included in this review. The studies assessed a median of 5 components of postural control (IQR, 3–6), with the Underlying Motor System being assessed across all studies. Then, Static Stability was the most frequently assessed component, evaluated in 88.2% of the studies (n=239). Dynamic Stability was assessed in 76.0% of the studies (n=206), while Anticipatory Postural Control was assessed in 75.3% of the studies (n=204). Cognitive Influences and Verticality were the least frequently assessed components, with 13.6% of the studies (n=37) for the former and only 6.3% of the studies (n=17) for the latter.
DISCUSSION
This scoping review focused on the standardized measures used in contemporary research to assess postural control among older adults. With a total of 271 articles, this review has summarized a significant amount of research and attempted to provide a clear overview of the topic over the last decade. This review reports on a considerable number of measures used (n=49) and the wide variety of tests used across studies complicates the synthesis of results, making it difficult to compare findings directly, increases variability in outcomes, and limits the consistency of assessment standards, which can undermine the robustness and reliability of conclusions. The most frequently standardized measure used to assess postural control was the TUG, followed by the BBS, the Static force platform eyes open test, and the Single-leg test. Furthermore, components of postural control were heterogeneously assessed, ranging from constantly (for Underlying Motor Systems) to very rarely (for Reactive Postural Control, Cognitive Influences, and Verticality) which highlights poor coverage of components and suggest major gaps in the current literature.
The TUG was the most frequently employed measure for the assessment of postural control in RCTs designed to enhance balance in older adults, consistently with other systematic reviews focusing on clinical population.14,21) TUG is also widely used by rehabilitation clinicians across the world,22,23) due to its simplicity, quickness and easiness to administer.24) In addition, the psychometric data for this test is of a high quality when applied to older people.25,26) However, the postural control components coverage of the TUG is low, with only three components (Underlying Motor Systems, Anticipatory Postural Control, and Dynamic Stability) assessed. Our review found that the second most used measure in research with older people was the BBS, a more complex and time-consuming measure.27) Like the TUG, previous studies and reviews have shown that the BBS is widely used by both researchers and clinicians in different areas of rehabilitation sciences14,21-23) and presents good psychometric qualities.28) However, it is noteworthy that the TUG and BBS are markedly distinct in terms of ease of use and the components of postural control they encompass, with the BBS assessing six out of nine components of postural control.9) Table 2 provides a summary of the psychometric qualities, component coverage, duration and equipment requirements of the TUG and BBS, comparing them with Balance Evaluation Systems Test (BESTest), which is the only measure available with full component coverage.9)
This review also showed that the components of postural control assessed in each study varied widely (from 6.3% of included studies for Verticality to 100% for Underlying Motor Systems). Thus, some components were infrequently assessed (Reactive Postural Control in 15.1% of included studies, Cognitive Influences in 13.6%, and Verticality in 6.3%), although these components make an important contribution to postural control performance and fall avoidance.29-31) To date, there is no comparable review assessing components coverage of postural control measures in older adults, preventing any direct comparison with the current work. However, in a similar approach, in a 2013 review, Pardasaney et al.32) realized a content analysis of balance measures for older adults, using a different conceptual model (i.e., Gentile’s model). The authors highlighted the plethora of available balance measures, the important content gaps in measures, global lack of representation in some content areas and limited incorporation of environmental manipulations. More than 10 years later, our results are in line with those of Pardasaney et al.32)
In addition, many previous studies have shown that the associations between measures with high coverage of postural control components and measures with low component coverage vary considerably.33,34) Consequently, these measures cannot be regarded as equivalent for substitution. This evidence suggests that these measures assess postural control in disparate ways, with the choice of measure having a significant impact on study outcomes due to the degree of coverage. As clinical guidelines and public health policies are indirectly based on high-quality research,35,36) low conceptual coverage of postural control in studies could have implications for clinical practice and policy decisions. To date, the BESTest, created by Horak et al.8) in 2009, as described in Table 2, is the only measure that encompasses all nine components of postural control.9) Despite the limited number of studies that have utilized this test in this review (1% of the included studies) may be due to its time constraints,37) the inclusion of all nine components9) and its validated status33) in this population make this measure the preferred choice for future research, until the validation of a shorter alternative with a comparable components' coverage. Concerning clinical use, several factors have to be considered for a comprehensive and efficient assessment of postural control in older adults. These factors, along with future directions and additions are proposed in the Fig. 4.
Current factors to consider and future directions for choosing a measure for a comprehensive assessment of postural control in older adults in clinical practice. Current factors include coverage of all the components of postural control, proven high psychometric qualities, adaptation to the equipment and time available, as well as to the individual's uniqueness and values. Future directions could include environmental variations (e.g. changes in lighting, interactions with people/objects), strong ecological validity (by approximating real-life conditions) and the inclusion of new technologies (e.g., using immersive virtual reality). This figure has been created with Canva (Canva Pty Ltd).
Limitations
Although the study conforms to JBI recommendations, it is important to note the absence of a critical appraisal of the sources utilized. It should be acknowledged, however, that this is not a mandatory requirement in the JBI recommendations. Despite the absence of critical appraisal, the review's compliance with other fundamental aspects of the recommendations provides a solid framework for the synthesis and interpretation of the collected data. Nonetheless, future research could be improved by incorporating critical appraisal techniques to enhance the rigor and validity of the findings. Another limitation concerns the categorization of postural control components for each assessment measure: the categorization was based on that carried out in the rigorous work by Sibley et al.9) and completed by consensus between the review team members if a new standardized measure was identified. However, this categorization remains subjective and open to interpretation, and we are aware that it may be questionable for certain components or measures. A last limitation is that particular phenotypes among older patients (e.g., frail people, people with cognitive impairments) were not taken account in analyses. These phenotypes could impact the suitability of several postural control measures included in this review, and future research is needed to clarify these impacts.
Conclusions
This scoping review systematically examined standardized measures used in RCTs to assess postural control in older adults, analyzing 271 studies. The TUG was the most frequently used, followed by the BBS, each differing in the postural control components they assess. In addition, our findings highlight significant variability in the assessment of these components outlined by the Postural Control Systems framework. Key components like Reactive Postural Control, Cognitive Influences, and Verticality were often underassessed, despite their importance. Our findings suggest the need for comprehensive assessment measures, such as the BESTest, which encompasses all nine components of postural control. Despite its limited use and time constraints, the BESTest is recommended for future research due to its thorough coverage and validated status. Yet, other tests should be investigated with shorter test durations, while still considering all the components of postural control. Future studies should prioritize such comprehensive measures to enhance the rigor and validity of findings, ultimately improving clinical guidelines and public health policies for older adults.
Notes
The authors would like to thank Dr.Carole Puil (IFPEK Rennes) and Dr.Nathalie Bonardet (IFPEK Rennes) for their expert input on postural control.
During the preparation of this work the author(s) used Rayyan (for assistance during the review screening process), ChatGPT and Deepl (as support for writing this article or the protocol for this review). After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
CONFLICT OF INTEREST
The researchers claim no conflicts of interest.
FUNDING
This work was financially supported by the association Institut de Formation en Pédicure-podologie, Ergothérapie et Kinésithérapie (IFPEK) Rennes.
AUTHOR CONTRIBUTIONS
Conceptualization, RP, BL, KJ, JE, KMS; Data curation, RP, BL, KJ; Investigation, RP, BL, KJ; Methodology, RP, BL, KJ; Supervision, RP, BL, KJ; Formal analysis, RP, BL, KJ; Writing_original draft, RP, BL, KJ, JE, KMS; Writing_review & editing, RP, BL, KJ, JE, KMS.
SUPPLEMENTARY MATERIALS
Supplementary materials can be found via https://doi.org/10.4235/agmr.25.0019.
Postural control components definitions
Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist
Search strategies
Charting table
Detailed characteristics of the 271 included studies
Distribution by country of included studies
Full list of postural control measures used in included studies
Postural control components coverage in measures used in included studies
