DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY
REVIEW
Psychometric properties of functional balance tests in children: a literature review EVI VERBECQUE 1
| PAULA HENTSCHEL LOBO DA COSTA 2,3 | LUC VEREECK 1 | ANN HALLEMANS 1,4
1 Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Science, University of Antwerp, Antwerp, Belgium. 2 Department of Physical Education, Federal University of S~ao Carlos, S~ao Carlos, Brazil. 3 Functional Morphology, Department of Biology, Faculty of Sciences, University of Antwerp, Antwerp; 4 Department of Translational Neurosciences, Faculty of Medicine and Health Science, University of Antwerp, Antwerp, Belgium. Correspondence to Evi Verbecque at Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Science, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk, Antwerp, Belgium. E-mail: [email protected]
PUBLICATION DATA
Accepted for publication 30th October 2014. Published online ABBREVIATIONS
BBS BBW FRT ICF-CY
OLH OLS PBS PRT TUDS TUG
Berg Balance Scale Balance Beam Walking Functional Reach Test International Classification of Functioning, Disability and Health, the Child and Youth Version One Leg Hopping One Leg Stance Pediatric Balance Scale Pediatric Reach Test Timed Up and Down Stairs Timed Up and Go
AIM Identifying balance problems are the first step towards monitoring and rehabilitation. Therefore, this paper aims to make an overview of the psychometric properties of the functional balance tests available for children. METHOD A literature search was performed in PubMED and Web of Science on 8 February 2014 and updated on 6 July 2014. A conceptual framework for functional balance tests was defined, taking balance control components and task constraints into account. The tests were selected for inclusion by consensus of 2-3 reviewers using the conceptual framework. RESULTS Fourteen tests were investigated in 25 articles and analysed within the conceptual framework. The Timed Up and Go test, Pediatric Balance Scale, and Pediatric Reach Test are well investigated and all show good reliability. Validity remains unclear because of lack of a criterion standard to measure balance control. INTERPRETATION Because of the lack of good methodological studies, strong evidence for the use of one or more functional balance tests in children cannot be provided. Moreover, it is necessary that a criterion standard to measure balance is established.
The concept of balance has been of great interest for many researchers over the past 20 years.1–8 Nevertheless, balance is defined in many different ways and a universal definition of balance is still lacking in literature.1–8 Huxham et al.2 describe balance as the ‘foundation for all voluntary motor skills’, and therefore an integral component of functioning. In their point of view, balance is the umbrella that comprises the control of posture (e.g. maintaining a posture) and the control of equilibrium (e.g. the control of destabilizing forces acting upon the body). The combination of both is a necessity to ensure stability of the body during widely differing motor tasks allowing skilled movement.2 This definition of balance, provided from a functional point of view, closely matches the biomechanical requirements of balance as described by Winter.3 Normal balance control thus requires both postural control and equilibrium control.2 Postural control can be defined as achieving a desired body position (e.g. upright standing) and maintaining this position in any static (maintaining a posture) or dynamic (performing a motor skill) situation.1–8 Equilibrium control relates to maintaining intersegmental © 2014 Mac Keith Press
stability of the body and its parts despite the gravitational and inertial forces acting on it.7 Adequate reactions to destabilizing forces are required to prevent a fall.2–8 When destabilizing forces are large, actions (balance strategies) are required to restore balance (e.g. the stepping strategy). This requires perception of sensory information and central processing involving the selection of an appropriate balance strategy leading to an adequate motor output.1,2,8 In summary, Huxham et al.2 state that the concept of balance control is the combination of proactive, predictive, and reactive mechanisms that relate to achieving, maintaining, and restoring balance, respectively. In clinical practice, balance control can be evaluated by investigating the motor output (e.g. task performance). But balance control, and thus task performance, is influenced by task, environmental, and individual constraints.1,2,8 To correctly assess an individual’s balance, these constraints should be taken into consideration. On the other hand, a test should be standardized, making it difficult to address the environmental factors. Task constraints can be taken into consideration1 and can be classified accordingly: DOI: 10.1111/dmcn.12657 1
(1) stability or maintaining a posture refers to the ability to keep the body balanced in a static position such as bipedal stance, tandem stance, or one leg stance (static body stability);9 (2) quasi-mobility or moving within a posture refers to the ability to keep the body balanced during movements in one posture, for example reaching forward while standing, kicking a ball (dynamic body stability), or transfers between postures such as going from sitting to standing (transfer stability);9 (3) mobility or moving relative to the base of support refers to the ability to keep the body balanced during locomotion (stability during locomotion).9 Depending on their age, children adopt different balance strategies determined not only by the difficulty of the task (task constraints as defined above), but also by the characteristics of each developmental period.1,8,10 Regarding static body stability, segmental control develops in a cephalo-caudal fashion, with the infant first achieving control over the head, then the trunk and extremities.1,8 The ability to control these body segments serves as a prerequisite for adequate postural control in standing.1,8 For stability during locomotion, on the other hand, the segmental organization of postural control, meaning the ability to dissociate movements of pelvis, trunk, and head, develops in a caudo-cephalic fashion.10 Assaiante et al.10 showed that children younger than 6 years of age, will use an ‘en bloc’ strategy to perform walking-like tasks, stabilizing the pelvis but not yet the head. Children from 6 to 7 years of age achieve head control in space during locomotion, but not in more difficult tasks (e.g. narrow beam walking), as this is only achieved at 7 to 8 years of age.10 In children with neurological (both central and peripheral), orthopaedic, and/or vestibular disorders, balance deficits are a common problem.6,11–15 For example, children with sensorineural hearing loss perform significantly poorer on the balance subtest of the Bruininks-OseretskyTest, 2nd edition, in comparison to their typically developing peers.15 These balance deficits need to be detected early in childhood because of their potential impact on motor development. Indeed, balance control plays an important role in overall development. It is essential for both acquiring and performing simple and complex gross and fine motor skills.6,7,12,13 Identifying possible balance problems is the first step towards monitoring and rehabilitation. This highlights the need for an adequate assessment tool for balance control in children. In clinical practice, numerous tests for measuring balance in children are available; the tests can be classified as developmental scales with balance subscales (e.g. balance subtest of the Bruininks-Oseretsky-Test, 2nd edition) or exclusive balance measurement tools with either functional balance tests (e.g. the Timed Up and Go test) or technical tests (e.g. posturography). According to Saether, ‘a good assessment tool should address the domain of concern (validity), be easily administered (especially with small children), be reliable in the population of interest, and be responsive to change’.6 Developmental scales are not designed to specifically address balance control and 2 Developmental Medicine & Child Neurology 2014
• •
What this paper adds A conceptual framework for functional balance tests in children. An overview of the psychometrics investigated in 14 functional balance tests.
technical tests are not easy to administer in clinical practice. The functional balance tests best fit these criteria and therefore will be the focus of this literature review. An overview of balance measurement tools and functional balance tests investigated in the literature and their psychometric properties available for children could help clinicians in choosing the appropriate test for the child or population under consideration. Even though trunk control is a prerequisite for balance in more upright postures, it is not within the scope of this review to investigate trunk control per se. The International Classification of Functioning, Disability and Health, the Child and Youth Version (ICF-CY) describes ‘control of voluntary movement functions’ clearly as a function, therefore trunk control can be classified as a (body) function as it implies control over and coordination of voluntary movements of the trunk.16 The focus of this review, however, will be on activity levels of the ICF-CY (e.g. changing and maintaining body position, maintaining a position, transferring oneself, and so on).16 Higher upright postures such as the motor milestones of independent standing and independent walking are achieved in typically developing children between 12 months and 18 months of age. Therefore this review will focus on the age of 18 months and older. This cut-off is chosen because it implies that children of that age are able to sit, stand, and walk independently and thus can be assessed in a large variety of tasks at activity levels of the ICF. In summary, through a literature review this paper will address the following research question: What are the functional balance tests available for children between 18 months and 18 years of age that have been investigated regarding their psychometric properties such as reliability, validity, and responsiveness?
METHOD Data sources and searches A literature search was performed in PubMED and Web of Science on 8 February 2014 and updated on 6 July 2014 using both free-text terms and controlled terminology (MeSH) (Table SI, supporting information published online). Screening was performed independently by two authors (EV, PHLC). In the first phase, the selection criteria (Table SI) were applied on title and abstract, and in the second phase on the full text. An inventory was made of the reason for exclusion. In case of disagreement, consensus was sought or a third author (AH) was consulted. In addition, reference lists of selected articles were screened. Data extraction Data extraction, performed in duplicate (EV, PHLC), was based upon the following variables: age of the study population, developmental characteristics or disorders, sample
with balance beam walking and one leg stance eyes open and eyes closed in typically developing children.17 One leg stance (OLS)17–20,31 has been investigated in typically developing children,17–20,31 and children with hearing impairment,19,20 and children with cerebral palsy (CP).31 OLS seems to be a highly reliable test (Table II). Results with regard to concurrent validity were conflicting: De Kegel et al.19 found moderate to very strong correlations with one legged hopping and balance beam walking (Table III), whereas Humphriss et al.17 showed no correlations between OLS and tandem stance on a balance beam, walking on a balance beam and tandem gait. Functional Reach is defined as the maximum distance one can reach forward beyond arm’s length while maintaining a fixed base of support in the standing position (Table I).21 The psychometrics of the Functional Reach Test (FRT) were assessed in children with CP and typical development.21,22,25,31 The Pediatric Reach Test (PRT) is an adapted version of the FRT especially designed for children and has been investigated in children with typical development,27,28 children with traumatic brain injury,23,24 and children with hearing impairment.26 Norris et al. provided FRT cut-off values for children 3 to 5 years old, whereas Deschmukh et al. provided PRT reference values for 6- to 12-year-old children.25,28 Excellent results for reliability were found for the FRT and PRT (Table II). Volkman et al.22 compared one-arm and two-arm reach and two measuring techniques (finger-to-finger and toe-to-finger distance) in
size, balance test under consideration, the psychometric property assessed (reliability, responsiveness, validity), and the time to perform the test.
RESULTS Study selection The literature search provided a total of 753 articles, of which 657 were unique hits. After selection, 18 articles were retained and through reference screening an additional seven articles were included, resulting in a net result of 25 articles (for the selection flowchart, see Fig. S1, supporting information published online). Balance measurement tools and their psychometric properties A total of 14 functional balance tests were identified that were investigated in children.17–40 These measurement tools were categorized as single balance measures17–34 or batteries of balance tests,34–40 and subsequently organized according to the conceptual framework we defined in the introduction (Table I). Eight single balance measures17–34 and six batteries of balance tests34–40 were identified. A detailed description of the test modalities for each test is available in Appendix S1 (supporting information published online). Single balance measures Tandem stance17 was investigated in typically developing children. Results from tandem stance did not correlate
Table I: Overview of the functional balance measurement tools available for children classified according to the constraints of the task (in relation to the support surface) and the balance control components Task constraints Quasi-mobility Stability Static body stability ‘maintaining’
One leg stance (eyes open/closed) Tandem stance (eyes open/closed)
Dynamic body stability
Transfer stability
Functional Reach Test Pediatric Functional Reach
One leg hopping Balance beam walking
Timed Up and Go
Berg Balance Scale Pediatric Balance Scale Ghent Developmental Balance Test Community Balance and Mobility Scale
Berg Balance Scale Pediatric Balance Scale Ghent Developmental Balance Test Community Balance and Mobility Scale
Mobility Stability during locomotion
Timed Up and Go Timed Up and Down Stairs Standardized Walking Obstacle Course
Berg Balance Scale Pediatric Balance Scale Ghent Developmental Balance Test Community Balance and Mobility Scale Dynamic Gait Index
Object interaction ‘achieving’ Obstacle negotiation ‘restoring’
Berg Balance Scale Pediatric Balance Scale Community Balance and Mobility Scale Dynamic Gait Index Standardizes Walking Obstacle Course
The constraints of the task in relation to the support surface comprise three conditions: stability, quasi-mobility, and mobility. Stability refers to static body stability, meaning the participant must stay in a certain position and thus maintain balance control either in a static or (quasi)dynamic situation. Quasi-mobility relates to balance control during movement between postures e.g. performing a transfer (transfer stability) or picking up an object from the floor (dynamic body stability). Mobility comprises stability during locomotion.
Review
3
Table II: Overview of the reliability of the functional balance measurement tools reported in literature n
Population
One leg stance (eyes open)
4920 2320 4731
One leg stance (eyes closed) Functional Reach Test
4920 2320
Pediatric Reach Test
One leg hopping Balance beam walking Timed Up and Go
5–12y 7–16y 8–14y
2424
Cerebral palsy Typical development Cerebral palsy, typical development Traumatic brain injury
2424
Typical development
7–14y
6526
Hearing impairment
6–11y
0.94–0.98 (F) 0.95–0.96 (L)
0.90 (F) 0.97 (L)
1027
Cerebral palsy
3–13y
0.54–0.88 (total)
0.50–0.93 (total)
49 2320 4920 2320 2621 5029
Typical development Hearing impairment Typical development Hearing impairment Cerebral palsy Typical development
6–12y 6–12y 6–12y 6–12y 5–12y 4–11y
2329
Developmental disabilities Traumatic brain injury Typical development Typical development
6–21y
Cerebral palsy, spina bifida Cerebral palsy, typical development Cerebral palsy Cerebral palsy, typical development Cerebral palsy Cerebral palsy Typical development Motor impairment Cerebral palsy
3–19y
Cerebral palsy Fetal alcohol spectrum disorders Typical development Typical development
8–13y 8–15y
Developmental disabilities Typical development
6–21y
Hearing impairment
1.5–6y
Acquired brain Injury
7–18y
2221 6922 4731
20
47
31
2032 4731 3021 2032 4034 2034 3636 1537 1039 39
Standardized Walking Obstacle Course Ghent Developmental Balance Test Community Balance and Mobility scale
Interrater reliability (ICC)
6–12y 6–12y 8–14y
4130
Dynamic Gait Index
Intrarater reliability (ICC)
Typical development Hearing impairment Cerebral palsy, typical development Typical development Hearing impairment
2424 2424 17630
Timed Up and Down Stairs Berg Balance Scale Pediatric Balance Scale
Age range
10 5029 2329 40
144
2240 32
41
0.99
0.98
0.83 0.94
SEM: 10.16s (9.9%) SEM: 8.71s (10%)
0.86 0.91
SEM: 13.37s (21.4%) SEM: 8.83s (28.9%)
0.95 0.39–0.93
0.98
7–14y
0.97 0.98 0.92 0.95 0.95 0.94
– 0.99 (time) 0.94–0.98 (steps)
(F) (L,P) (L,NP) (F) (L,P) (l, NP)
0.95 0.97 0.88 0.97 0.99 0.88–0.99 (time)
0.92–0.96 (steps) 0.93–0.99 (time) 0.94–0.99 (steps) 0.86 0.85 0.83 (1wk) 0.89 (same day) 0.99 (0.91–0.99) (same day)
7–14y 7–14y 3–9y
8–14y
0.99
0.99
6–14y 8–14y
0.99
0.99
5–12y 6–14y 5–7y 5–15y 6–13y
Measurement error
0.99
6–12y 6–12y
0.97
Test–retest reliability (ICC)
0.99
0.978–0.988
0.85*
8–15y 4–11y
0.998 0.94
0.997 0.905
1.00 0.999 0.850 0.998 0.958
0.91*, 0.98 0.82
0.71
0.99 (time) 0.94–0.97 (steps) 0.99 (time) 0.99 (steps)
1.5–6y
0.83–0.94 0.84–0.94 0.91–0.96 0.90–0.96 0.99
0.98 0.96
0.93, 0.95
SEM: 0.97 (F) (cm) SEM: 0.72 (L,P) (cm) SEM: 0.90 (L,NP) (cm) SEM: 1.41 (F) (cm) SEM: 0.80 (L,P) (cm) SEM: 0.97 (L,NP) (cm) Intrarater: SEM: 0.29–0.51 (F), SEM: 0.28–0.32 (L) (cm)
SEM: SEM: SEM: SEM:
4.21s 3.29s 6.74s 3.07s
(9%) (8.7%) (16.6%) (11.2%)
SEM: 0.60 (s) SEM: 0.23 (s)
REVIEW
Psychometric properties of functional balance tests in children: a literature review EVI VERBECQUE 1
| PAULA HENTSCHEL LOBO DA COSTA 2,3 | LUC VEREECK 1 | ANN HALLEMANS 1,4
1 Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Science, University of Antwerp, Antwerp, Belgium. 2 Department of Physical Education, Federal University of S~ao Carlos, S~ao Carlos, Brazil. 3 Functional Morphology, Department of Biology, Faculty of Sciences, University of Antwerp, Antwerp; 4 Department of Translational Neurosciences, Faculty of Medicine and Health Science, University of Antwerp, Antwerp, Belgium. Correspondence to Evi Verbecque at Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Science, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk, Antwerp, Belgium. E-mail: [email protected]
PUBLICATION DATA
Accepted for publication 30th October 2014. Published online ABBREVIATIONS
BBS BBW FRT ICF-CY
OLH OLS PBS PRT TUDS TUG
Berg Balance Scale Balance Beam Walking Functional Reach Test International Classification of Functioning, Disability and Health, the Child and Youth Version One Leg Hopping One Leg Stance Pediatric Balance Scale Pediatric Reach Test Timed Up and Down Stairs Timed Up and Go
AIM Identifying balance problems are the first step towards monitoring and rehabilitation. Therefore, this paper aims to make an overview of the psychometric properties of the functional balance tests available for children. METHOD A literature search was performed in PubMED and Web of Science on 8 February 2014 and updated on 6 July 2014. A conceptual framework for functional balance tests was defined, taking balance control components and task constraints into account. The tests were selected for inclusion by consensus of 2-3 reviewers using the conceptual framework. RESULTS Fourteen tests were investigated in 25 articles and analysed within the conceptual framework. The Timed Up and Go test, Pediatric Balance Scale, and Pediatric Reach Test are well investigated and all show good reliability. Validity remains unclear because of lack of a criterion standard to measure balance control. INTERPRETATION Because of the lack of good methodological studies, strong evidence for the use of one or more functional balance tests in children cannot be provided. Moreover, it is necessary that a criterion standard to measure balance is established.
The concept of balance has been of great interest for many researchers over the past 20 years.1–8 Nevertheless, balance is defined in many different ways and a universal definition of balance is still lacking in literature.1–8 Huxham et al.2 describe balance as the ‘foundation for all voluntary motor skills’, and therefore an integral component of functioning. In their point of view, balance is the umbrella that comprises the control of posture (e.g. maintaining a posture) and the control of equilibrium (e.g. the control of destabilizing forces acting upon the body). The combination of both is a necessity to ensure stability of the body during widely differing motor tasks allowing skilled movement.2 This definition of balance, provided from a functional point of view, closely matches the biomechanical requirements of balance as described by Winter.3 Normal balance control thus requires both postural control and equilibrium control.2 Postural control can be defined as achieving a desired body position (e.g. upright standing) and maintaining this position in any static (maintaining a posture) or dynamic (performing a motor skill) situation.1–8 Equilibrium control relates to maintaining intersegmental © 2014 Mac Keith Press
stability of the body and its parts despite the gravitational and inertial forces acting on it.7 Adequate reactions to destabilizing forces are required to prevent a fall.2–8 When destabilizing forces are large, actions (balance strategies) are required to restore balance (e.g. the stepping strategy). This requires perception of sensory information and central processing involving the selection of an appropriate balance strategy leading to an adequate motor output.1,2,8 In summary, Huxham et al.2 state that the concept of balance control is the combination of proactive, predictive, and reactive mechanisms that relate to achieving, maintaining, and restoring balance, respectively. In clinical practice, balance control can be evaluated by investigating the motor output (e.g. task performance). But balance control, and thus task performance, is influenced by task, environmental, and individual constraints.1,2,8 To correctly assess an individual’s balance, these constraints should be taken into consideration. On the other hand, a test should be standardized, making it difficult to address the environmental factors. Task constraints can be taken into consideration1 and can be classified accordingly: DOI: 10.1111/dmcn.12657 1
(1) stability or maintaining a posture refers to the ability to keep the body balanced in a static position such as bipedal stance, tandem stance, or one leg stance (static body stability);9 (2) quasi-mobility or moving within a posture refers to the ability to keep the body balanced during movements in one posture, for example reaching forward while standing, kicking a ball (dynamic body stability), or transfers between postures such as going from sitting to standing (transfer stability);9 (3) mobility or moving relative to the base of support refers to the ability to keep the body balanced during locomotion (stability during locomotion).9 Depending on their age, children adopt different balance strategies determined not only by the difficulty of the task (task constraints as defined above), but also by the characteristics of each developmental period.1,8,10 Regarding static body stability, segmental control develops in a cephalo-caudal fashion, with the infant first achieving control over the head, then the trunk and extremities.1,8 The ability to control these body segments serves as a prerequisite for adequate postural control in standing.1,8 For stability during locomotion, on the other hand, the segmental organization of postural control, meaning the ability to dissociate movements of pelvis, trunk, and head, develops in a caudo-cephalic fashion.10 Assaiante et al.10 showed that children younger than 6 years of age, will use an ‘en bloc’ strategy to perform walking-like tasks, stabilizing the pelvis but not yet the head. Children from 6 to 7 years of age achieve head control in space during locomotion, but not in more difficult tasks (e.g. narrow beam walking), as this is only achieved at 7 to 8 years of age.10 In children with neurological (both central and peripheral), orthopaedic, and/or vestibular disorders, balance deficits are a common problem.6,11–15 For example, children with sensorineural hearing loss perform significantly poorer on the balance subtest of the Bruininks-OseretskyTest, 2nd edition, in comparison to their typically developing peers.15 These balance deficits need to be detected early in childhood because of their potential impact on motor development. Indeed, balance control plays an important role in overall development. It is essential for both acquiring and performing simple and complex gross and fine motor skills.6,7,12,13 Identifying possible balance problems is the first step towards monitoring and rehabilitation. This highlights the need for an adequate assessment tool for balance control in children. In clinical practice, numerous tests for measuring balance in children are available; the tests can be classified as developmental scales with balance subscales (e.g. balance subtest of the Bruininks-Oseretsky-Test, 2nd edition) or exclusive balance measurement tools with either functional balance tests (e.g. the Timed Up and Go test) or technical tests (e.g. posturography). According to Saether, ‘a good assessment tool should address the domain of concern (validity), be easily administered (especially with small children), be reliable in the population of interest, and be responsive to change’.6 Developmental scales are not designed to specifically address balance control and 2 Developmental Medicine & Child Neurology 2014
• •
What this paper adds A conceptual framework for functional balance tests in children. An overview of the psychometrics investigated in 14 functional balance tests.
technical tests are not easy to administer in clinical practice. The functional balance tests best fit these criteria and therefore will be the focus of this literature review. An overview of balance measurement tools and functional balance tests investigated in the literature and their psychometric properties available for children could help clinicians in choosing the appropriate test for the child or population under consideration. Even though trunk control is a prerequisite for balance in more upright postures, it is not within the scope of this review to investigate trunk control per se. The International Classification of Functioning, Disability and Health, the Child and Youth Version (ICF-CY) describes ‘control of voluntary movement functions’ clearly as a function, therefore trunk control can be classified as a (body) function as it implies control over and coordination of voluntary movements of the trunk.16 The focus of this review, however, will be on activity levels of the ICF-CY (e.g. changing and maintaining body position, maintaining a position, transferring oneself, and so on).16 Higher upright postures such as the motor milestones of independent standing and independent walking are achieved in typically developing children between 12 months and 18 months of age. Therefore this review will focus on the age of 18 months and older. This cut-off is chosen because it implies that children of that age are able to sit, stand, and walk independently and thus can be assessed in a large variety of tasks at activity levels of the ICF. In summary, through a literature review this paper will address the following research question: What are the functional balance tests available for children between 18 months and 18 years of age that have been investigated regarding their psychometric properties such as reliability, validity, and responsiveness?
METHOD Data sources and searches A literature search was performed in PubMED and Web of Science on 8 February 2014 and updated on 6 July 2014 using both free-text terms and controlled terminology (MeSH) (Table SI, supporting information published online). Screening was performed independently by two authors (EV, PHLC). In the first phase, the selection criteria (Table SI) were applied on title and abstract, and in the second phase on the full text. An inventory was made of the reason for exclusion. In case of disagreement, consensus was sought or a third author (AH) was consulted. In addition, reference lists of selected articles were screened. Data extraction Data extraction, performed in duplicate (EV, PHLC), was based upon the following variables: age of the study population, developmental characteristics or disorders, sample
with balance beam walking and one leg stance eyes open and eyes closed in typically developing children.17 One leg stance (OLS)17–20,31 has been investigated in typically developing children,17–20,31 and children with hearing impairment,19,20 and children with cerebral palsy (CP).31 OLS seems to be a highly reliable test (Table II). Results with regard to concurrent validity were conflicting: De Kegel et al.19 found moderate to very strong correlations with one legged hopping and balance beam walking (Table III), whereas Humphriss et al.17 showed no correlations between OLS and tandem stance on a balance beam, walking on a balance beam and tandem gait. Functional Reach is defined as the maximum distance one can reach forward beyond arm’s length while maintaining a fixed base of support in the standing position (Table I).21 The psychometrics of the Functional Reach Test (FRT) were assessed in children with CP and typical development.21,22,25,31 The Pediatric Reach Test (PRT) is an adapted version of the FRT especially designed for children and has been investigated in children with typical development,27,28 children with traumatic brain injury,23,24 and children with hearing impairment.26 Norris et al. provided FRT cut-off values for children 3 to 5 years old, whereas Deschmukh et al. provided PRT reference values for 6- to 12-year-old children.25,28 Excellent results for reliability were found for the FRT and PRT (Table II). Volkman et al.22 compared one-arm and two-arm reach and two measuring techniques (finger-to-finger and toe-to-finger distance) in
size, balance test under consideration, the psychometric property assessed (reliability, responsiveness, validity), and the time to perform the test.
RESULTS Study selection The literature search provided a total of 753 articles, of which 657 were unique hits. After selection, 18 articles were retained and through reference screening an additional seven articles were included, resulting in a net result of 25 articles (for the selection flowchart, see Fig. S1, supporting information published online). Balance measurement tools and their psychometric properties A total of 14 functional balance tests were identified that were investigated in children.17–40 These measurement tools were categorized as single balance measures17–34 or batteries of balance tests,34–40 and subsequently organized according to the conceptual framework we defined in the introduction (Table I). Eight single balance measures17–34 and six batteries of balance tests34–40 were identified. A detailed description of the test modalities for each test is available in Appendix S1 (supporting information published online). Single balance measures Tandem stance17 was investigated in typically developing children. Results from tandem stance did not correlate
Table I: Overview of the functional balance measurement tools available for children classified according to the constraints of the task (in relation to the support surface) and the balance control components Task constraints Quasi-mobility Stability Static body stability ‘maintaining’
One leg stance (eyes open/closed) Tandem stance (eyes open/closed)
Dynamic body stability
Transfer stability
Functional Reach Test Pediatric Functional Reach
One leg hopping Balance beam walking
Timed Up and Go
Berg Balance Scale Pediatric Balance Scale Ghent Developmental Balance Test Community Balance and Mobility Scale
Berg Balance Scale Pediatric Balance Scale Ghent Developmental Balance Test Community Balance and Mobility Scale
Mobility Stability during locomotion
Timed Up and Go Timed Up and Down Stairs Standardized Walking Obstacle Course
Berg Balance Scale Pediatric Balance Scale Ghent Developmental Balance Test Community Balance and Mobility Scale Dynamic Gait Index
Object interaction ‘achieving’ Obstacle negotiation ‘restoring’
Berg Balance Scale Pediatric Balance Scale Community Balance and Mobility Scale Dynamic Gait Index Standardizes Walking Obstacle Course
The constraints of the task in relation to the support surface comprise three conditions: stability, quasi-mobility, and mobility. Stability refers to static body stability, meaning the participant must stay in a certain position and thus maintain balance control either in a static or (quasi)dynamic situation. Quasi-mobility relates to balance control during movement between postures e.g. performing a transfer (transfer stability) or picking up an object from the floor (dynamic body stability). Mobility comprises stability during locomotion.
Review
3
Table II: Overview of the reliability of the functional balance measurement tools reported in literature n
Population
One leg stance (eyes open)
4920 2320 4731
One leg stance (eyes closed) Functional Reach Test
4920 2320
Pediatric Reach Test
One leg hopping Balance beam walking Timed Up and Go
5–12y 7–16y 8–14y
2424
Cerebral palsy Typical development Cerebral palsy, typical development Traumatic brain injury
2424
Typical development
7–14y
6526
Hearing impairment
6–11y
0.94–0.98 (F) 0.95–0.96 (L)
0.90 (F) 0.97 (L)
1027
Cerebral palsy
3–13y
0.54–0.88 (total)
0.50–0.93 (total)
49 2320 4920 2320 2621 5029
Typical development Hearing impairment Typical development Hearing impairment Cerebral palsy Typical development
6–12y 6–12y 6–12y 6–12y 5–12y 4–11y
2329
Developmental disabilities Traumatic brain injury Typical development Typical development
6–21y
Cerebral palsy, spina bifida Cerebral palsy, typical development Cerebral palsy Cerebral palsy, typical development Cerebral palsy Cerebral palsy Typical development Motor impairment Cerebral palsy
3–19y
Cerebral palsy Fetal alcohol spectrum disorders Typical development Typical development
8–13y 8–15y
Developmental disabilities Typical development
6–21y
Hearing impairment
1.5–6y
Acquired brain Injury
7–18y
2221 6922 4731
20
47
31
2032 4731 3021 2032 4034 2034 3636 1537 1039 39
Standardized Walking Obstacle Course Ghent Developmental Balance Test Community Balance and Mobility scale
Interrater reliability (ICC)
6–12y 6–12y 8–14y
4130
Dynamic Gait Index
Intrarater reliability (ICC)
Typical development Hearing impairment Cerebral palsy, typical development Typical development Hearing impairment
2424 2424 17630
Timed Up and Down Stairs Berg Balance Scale Pediatric Balance Scale
Age range
10 5029 2329 40
144
2240 32
41
0.99
0.98
0.83 0.94
SEM: 10.16s (9.9%) SEM: 8.71s (10%)
0.86 0.91
SEM: 13.37s (21.4%) SEM: 8.83s (28.9%)
0.95 0.39–0.93
0.98
7–14y
0.97 0.98 0.92 0.95 0.95 0.94
– 0.99 (time) 0.94–0.98 (steps)
(F) (L,P) (L,NP) (F) (L,P) (l, NP)
0.95 0.97 0.88 0.97 0.99 0.88–0.99 (time)
0.92–0.96 (steps) 0.93–0.99 (time) 0.94–0.99 (steps) 0.86 0.85 0.83 (1wk) 0.89 (same day) 0.99 (0.91–0.99) (same day)
7–14y 7–14y 3–9y
8–14y
0.99
0.99
6–14y 8–14y
0.99
0.99
5–12y 6–14y 5–7y 5–15y 6–13y
Measurement error
0.99
6–12y 6–12y
0.97
Test–retest reliability (ICC)
0.99
0.978–0.988
0.85*
8–15y 4–11y
0.998 0.94
0.997 0.905
1.00 0.999 0.850 0.998 0.958
0.91*, 0.98 0.82
0.71
0.99 (time) 0.94–0.97 (steps) 0.99 (time) 0.99 (steps)
1.5–6y
0.83–0.94 0.84–0.94 0.91–0.96 0.90–0.96 0.99
0.98 0.96
0.93, 0.95
SEM: 0.97 (F) (cm) SEM: 0.72 (L,P) (cm) SEM: 0.90 (L,NP) (cm) SEM: 1.41 (F) (cm) SEM: 0.80 (L,P) (cm) SEM: 0.97 (L,NP) (cm) Intrarater: SEM: 0.29–0.51 (F), SEM: 0.28–0.32 (L) (cm)
SEM: SEM: SEM: SEM:
4.21s 3.29s 6.74s 3.07s
(9%) (8.7%) (16.6%) (11.2%)
SEM: 0.60 (s) SEM: 0.23 (s)
