If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Portegijs E, Edgren J, Salpakoski A, Kallinen M, Rantanen T, Alen M, Kiviranta I, Sihvonen S, Sipilä S. Balance confidence was associated with mobility and balance performance in older people with fall-related hip fracture: a cross-sectional study.
To study the relationship between balance confidence, a concept closely related to fear of falling, mobility and balance performance, and perceived mobility limitation in older people after a fall-related hip fracture.
Cross-sectional analyses of pretrial data of 2 randomized controlled trials of physical rehabilitation.
University research center.
Community-dwelling people aged over 60 years, 6 weeks to 7.5 years after a fall-related hip fracture (N=130).
Main Outcome Measures
The main outcome was the self-reported Activities-specific Balance Confidence (ABC) scale score. Assessments also included perceived ability to walk outdoors or climb 1 flight of stairs, and assessments of self-preferred walking speed, modified Timed-Up-and-Go test, and Berg Balance Scale.
Higher ABC scale scores were related to better mobility and balance performance (ρ>.47) and perceived mobility function (ρ>.54). In univariate general linear models, all associations also remained significant after adjustment for age, sex, time since fracture, number of chronic diseases, and either level of physical activity or muscle strength of the fractured leg. An ABC scale score <85 points identified those with mobility and balance limitation across measures.
In people who have had a fall-related hip fracture, an independent relationship exists between balance confidence and mobility and balance performance as well as perceived mobility function. Since lack of balance confidence may compromise rehabilitation and recovery, the ABC scale may help to identify older hip fracture patients with mobility and balance limitation.
may exist among older people with or without a history of falls. However, those with experience of a fall or fall-related trauma are likely to adapt their behavior because of fear of a new fall. After hip fracture, older people often experience lack of balance confidence, which may be a correct appraisal of their increased risk for falls.
Balance confidence may contribute to the lack of functional recovery after hip fracture. Lack of balance confidence and fear of falling are reported to have a debilitating effect on mobility and functioning in geriatric rehabilitation patients
Our aim was to examine the relationship between performance-based and self-reported measures of mobility and balance function in older people with a previous fall-related hip fracture. In addition, the study aimed to determine whether this relationship remained after adjustment, including potential underlying mechanisms, such as reduced physical activity level and poor muscle strength. Finally, we searched for a cutoff value for the ABC scale to identify those with mobility and balance limitation across the different measures.
This study was a joint analysis of 2 randomized controlled trials investigating health, functional capacity, and rehabilitation of people with a history of hip fracture.
Both studies were performed at the same research center, using the same equipment, and participants were recruited from the same health care district with identical inclusion and exclusion criteria. Data of these studies were pooled to increase sample size. Only the pretrial data are used in this cross-sectional study.
The methods of both trials have been described before.
Patient records at the Central Finland Central Hospital were reviewed (in the fall of 2004–2005 and throughout 2008–2010) to recruit community-dwelling people older than 60 years who had been operated on for femoral neck or trochanteric fracture. All potential patients were informed about the study (N=748). Those willing to participate were interviewed over the telephone or met during the inpatient period at the health care center to ensure suitability (N=268). Exclusion criteria were inability to move outdoors without assistance from another person, amputation of a lower limb, severe progressive or neurologic diseases, alcoholism, and severe memory problems (Mini-Mental State Examination score<19).
Of the 149 eligible patients, 130 were included in the present analyses on the basis of an additional criterion of having experienced a hip fracture due to a low-impact fall, which may exacerbate the perception of lack of balance confidence. The ethical committee of the local health care district approved both of the study protocols. Participants gave their written informed consent prior to the assessments. The assessments were performed 6 weeks to 7.5 years after the fracture.
A physician and research nurse performed a thorough clinical examination to assess general health status. Details of the fracture and repair (internal fixation vs. arthroplasty) and the number of chronic conditions (present for at least 3 months) were confirmed according to a questionnaire and medical records. Time since fracture was calculated as the number of days between the date of fracture and the date of assessments.
Data collection of the following measures was performed by trained research assistants. Balance confidence in carrying out activities without becoming unsteady was assessed by interview using a modified Finnish version of the ABC scale.
In this modified version, items 14 and 15 regarding escalators in the original scale were replaced with 2 items on “riding on a bicycle in a street with light traffic/in a heavily trafficked street with no bike path.” Bicycling was a more relevant activity for Finnish older people because large malls with escalators were very uncommon until recently. Subjects were asked to report their confidence levels when carrying out 16 activities, including those performed outdoors. Each activity was rated from 1 (no confidence) to 10 (total confidence); total score ranged from 16 to 160. The ABC scale was reported to be valid and reliable in older populations.
which evaluates the ability to perform 14 tasks related to the subject's skills such as to sit down, stand up, reach forward, turn 360 degrees, and stand on 1 leg. Each task was rated from 0 (incapable) to 4 (safe and independent); total score ranged from 0 to 56. BBS had high intra- and interrater reliability.
Participants were allowed to use their assistive device commonly used for walking indoors during mobility tests. Self-preferred walking speed (m/s) was calculated from the shortest time to walk 10 meters, assessed using photocells.a Three meters was allowed for acceleration and deceleration. This test has been shown to be valid and reliable.
Self-reported mobility was assessed using a questionnaire. Participants were asked about perceived difficulty to walk outdoors and to climb 1 flight of stairs. The response categories were as follows: (1) no difficulty, (2) some difficulty, (3) major difficulty, (4) unable without help from another person, and (5) unable even with help. Because of low frequency in categories 3 to 5, they were joined for the analyses (“major difficulty or unable”).
with slight modifications. The highest category of the initial scale was divided into 2 categories, separating those participating in regular exercise fitness activities from those active in competitive sports. The 7-point scale ranged from 1 (mostly sitting) to 7 (participation in competitive sports). Categories 4 to 7 were combined because of low frequencies, thus leaving 4 groups: (1) mostly sitting, (2) light physical activity, such as light household tasks; (3) moderate physical activity for less than 3 hours a week, such as walking longer distance and domestic work; and (4) moderate activity for more than 4 hours a week and/or more strenuous activity multiple times a week.
Maximal voluntary isometric knee extension strength of the fractured leg was assessed using an adjustable dynamometer chair.b The ankle was attached to a strain-gauge system with the knee angle fixed at 60 degrees from full extension. Participants were encouraged to extend the leg as forcefully as possible. After 2 to 3 practice trials, the highest force of at least 3 measurements was used for analysis. Each contraction was maintained for 2 to 3 seconds. The test has been shown to be valid and reliable.
Participants with missing variables in knee extension strength (n=15), walking (n=8), TUG (n=13), and BBS (n=6) tests were dropped from the respective analysis only. BBS score, walking speed, and TUG test values were analyzed as continuous as well as categorical variables on the basis of cutoff points used to predict falls.
Variable distribution was tested with Kolmogorov-Smirnov tests. Group differences were tested with independent t tests and χ2 tests. Spearman (ρ) and Pearson (r) correlation coefficients were calculated for relationships between ABC scale score, measures of mobility and balance performance, and perceived mobility limitation as well as confounders. Fracture repair type was not significantly (P>.05) associated with any mobility or balance measure and was therefore not included in multivariable analyses. Univariate general linear models were used to compare groups on the basis of categorized mobility and balance performance measures and perceived mobility limitation. Each model included ABC scale score (crude model) and was adjusted for (1) age, sex, time since hip fracture, and number of chronic diseases and in addition either (2) level of physical activity or (3) knee extension strength of the fractured limb. Separate receiver operating characteristic curves were drawn for the ABC scale score to identify those with poorer mobility and balance performance, or major difficulty in perceived mobility measures. For each measure, the optimal cutoff point(s) (highest sensitivity and specificity) was determined. These cutoff points were then used in sensitivity and specificity analyses to choose the cutoff point most suitable to identify those with mobility and balance limitation across outcomes. Finally, the cutoff score identified was used to predict mobility and balance limitation in logistic regression analyses. Odds ratios and 95% confidence intervals of crude (unadjusted) models were reported. PASW Statistics 18c was used for the statistical analyses, and statistical significance was set at P≤.05.
Table 1 shows the participants' characteristics. Their average age ± SD was 77.6±7.2 years, and the majority were women. On average, 1.5±1.9 years had passed since the hip fracture.
Table 1Means ± SD or Prevalence of the Participant Characteristics and Spearman Correlation Coefficients of Each Measure With the ABC Scale
Correlation coefficients between the ABC scale score and continuous mobility and balance performance measures indicated that better performance (higher score; for TUG test lower score) was associated with higher balance confidence (higher ABC scale score) (see table 1). Similar correlations were found between the ABC scale score and categorized performance-based measures; ρ=.60 (P<.001) for BBS, ρ=.48 (P<.001) for TUG test, and ρ=−.47 (P<.001) for walking speed, respectively. For the self-reported measures, less difficulty was related to higher balance confidence; the correlation coefficients with the ABC scale score were ρ=−.54 (P<.001) for ability to walk outdoors and ρ=−.57 (P<.001) for stair climb ability, respectively.
Group differences were calculated for all confounding variables (data not shown). Those with poorer walking speed, TUG test score, and BBS score and those with perceived mobility limitation were significantly older (P≤.014) and had lower levels of physical activity (P<.001) and muscle strength (P≤.005) than those with better functioning. For those with poorer walking speed, TUG test scores, and perceived limitation to walk outdoors, significantly less time had passed since the hip fracture (P≤.041). Those with poorer TUG test scores and BBS scores and those with perceived mobility limitation had a higher number of chronic diseases (P≤.029) than did those with better functioning. Lower ABC scale score correlated significantly with higher age (r=.37, P<.001) and number of diseases (ρ=−.29, P=.001) and lower level of physical activity (ρ=.47, P<.001) and muscle strength (ρ=.40, P<.001) (see table 1).
Figure 1 shows a clear gradient of decreasing ABC scale scores in groups with poorer mobility and balance performance and perceived mobility limitation. Group differences also remained significant (P<.026) after adjustment for age, sex, time since hip fracture, and the number of chronic diseases, as well as after additional adjustment for level of physical activity or knee extension strength in the fractured limb (table 2). Level of physical activity and knee extension strength attenuated the relationship between the ABC scale score and all mobility and balance measures; however, the relationships remained significant.
Table 2Differences in Mean ABC Scale Tested With Adjusted Univariate Models for the Mobility and Balance Performance Groups Based on Previously Reported Cutoff Values and the Groups With Perceived Mobility Limitation
Receiver operating characteristic curves of each mobility and balance performance or perceived mobility limitation measure with the ABC scale score suggested several potential cutoff points: 68.5 (ability to walk outdoors and walking speed), 76.5 (walking speed and TUG test), 78 (ability to walk outdoors and climb stairs), 80.5 (walking speed), 84.5 (ability to climb stairs and TUG test), and 85.5 (BBS). All were used in the following sensitivity and specificity analyses to identify those with poorer mobility and balance performance and perceived mobility limitation for each measure. Using 84.5 points for cutoff rendered good sensitivity (≥.73) and specificity (≥.70) across all measures (table 3). In logistic regression analyses, those with an ABC scale sum score <85 points had an odds ratio of 18.7 (95% confidence interval 6.0–58.0) for having major outdoors walking difficulty and 11.7 (4.6–29.9) for major stair climb difficulty. For the performance-based tests, the odds ratios were 12.6 (5.3–29.8) for BBS, 7.3 (3.0–17.8) for the TUG test, and 6.3 (2.6–15.0) for walking speed, respectively.
Table 3Sensitivity and Specificity Analyses Using 85 Points as Cutoff Value for the ABC Scale (to Identify Those With Poorer Mobility and Balance Performance or Perceived Mobility Limitation)
Our study shows that balance confidence was associated with a range of measures of mobility and balance performance and perceived mobility limitation in older people after a fall-related hip fracture. The relationships found also remained significant after considerable adjustment. An ABC scale score <85 points identified most participants with mobility and balance limitation across the different measures. The ABC scale may be useful for clinicians to identify hip fracture patients with or at risk for mobility and balance limitation. Addressing lack of balance confidence together with improving mobility and balance performance by physical rehabilitation may positively affect the potential for functional recovery of hip fracture patients.
In relatively healthy populations of older people, a reduced ABC scale score, indicating lack of confidence to perform mobility tasks without loss of balance, has been associated with reduced performance as measured in walking speed,
The mean ABC scale score in our study, when converted to a scale from 0% to 100% (being 55), was similar as the means in 2 studies of patients about 4 months after hip fracture (being 59 and 61, respectively).
Predicting falls within the elderly community: comparison of postural sway, reaction time, the Berg balance scale and the Activities-specific Balance Confidence (ABC) scale for comparing fallers and non-fallers.
have previously determined a cutoff score for the ABC scale. In their study, a score of less than 67% (ie, 107 points) on the ABC scale increased the risk for falls in relatively healthy older people. Our population with a previous fall-related hip fracture had lower balance confidence (two thirds scored below 107 points). Therefore, we determined a new cutoff point to identify persons with mobility and balance limitation after a fall-related hip fracture. Those with an ABC scale score <85 points had a 6 to 18 times increased risk of having poor mobility and balance performance or perceived mobility limitation than those with higher ABC scale scores. The association between balance confidence, mobility and balance performance, or perceived mobility limitation was independent of time since fracture. This may support the hypothesis that poor balance confidence may implicate poorer recovery potential from an acute event
Identifying older hip fracture patients with low balance confidence and at risk for mobility and balance limitation may be clinically relevant. In our study, the ABC scale had an independent association with all measures of mobility and balance performance and perceived mobility limitation. Early identification of hip fracture patients at risk may also have an impact on other health outcomes, such as new falls, disability, and loss of independence. The ABC scale may also be suitable in the time-pressured clinical practice.
Early physical rehabilitation, including progressive resistance training or other physical exercises, may prevent or reduce the major functional decline associated with hip fracture.
intervention aiming to improve physical function may also improve balance confidence. In hip fracture patients attending inpatient rehabilitation, no relationship between change in the ABC scale score and change in physical function was observed
However, systematic reviews in older hip fracture patients of multidisciplinary interventions and interventions aiming to improve both physical and psychological function were not able to demonstrate better outcomes when compared with regular care after hip fracture.
Because of our inclusion criteria (community-dwelling, being able to come to our research center for measurements, being able to walk outdoors independently) participants were relatively well-functioning older people compared with hip fracture patients in general. Generalization of the results should be done with caution. We included older people with a large time range since the fall-related hip fracture (6wk to 7.5y). The time since fracture was related to performance-based mobility and balance measures; however, it did not affect the relationship between the ABC scale score, mobility and balance performance, or perceived mobility limitation. This may suggest that lack of balance confidence persistently affects mobility and balance performance in older people with a history of fall-related hip fracture. The sample size in this study allowed for considerable adjustment. We are therefore confident that the relationship between balance confidence and the different measures of mobility and balance performance and perceived mobility limitation was independent. Because of the cross-sectional study design, the chronological order of lack of balance confidence and limitations in mobility and balance performance and their relationship with the hip fracture event remain unclear. Longitudinal studies are needed to confirm associations and determine cause-effect relationships.
In older people with a fall-related hip fracture, an independent relationship exists between balance confidence and a range of performance-based and self-reported mobility and balance performance measures. In this group of older people, a score of <85 points on the ABC scale identified those with mobility and balance limitation. Identification of persons with lack of balance confidence seems clinically relevant as it may compromise functional recovery from the hip fracture. Potentially, rehabilitation may be more effective, when lack of balance confidence is taken into account or targeted. However, further study is needed to develop effective strategies to improve balance confidence and reduce the functional decline associated with hip fracture.
Predicting falls within the elderly community: comparison of postural sway, reaction time, the Berg balance scale and the Activities-specific Balance Confidence (ABC) scale for comparing fallers and non-fallers.
Current affiliations: Department of Medical Rehabilitation, Oulu University Hospital, and Institute of Health Sciences, University of Oulu, Oulu, Finland (Alen); Department of Orthopaedics and Traumatology, University of Helsinki, and Helsinki University Central Hospital, Helsinki, Finland (Kiviranta); and School of Health and Social Studies, JAMK University of Applied Sciences, Jyväskylä, Finland (Sihvonen).
Supported by Finnish Ministry of Education and Culture, Kela–The Social Insurance Institution of Finland, Juho Vainio Foundation, and Finnish Cultural Foundation. The funders had no further interest or involvement in the design, methods, subject recruitment, data collection, analysis, and preparation of the manuscript. Gerontology Research Center is a joint effort between the University of Jyväskylä and the University of Tampere.
Clinical Trial Registration Nos.: ISRCTN34271567 and ISRCTN53680197.
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.