According to the NIH, more than one-third of adults over the age of 65 fall each year. Over 1.6 million seniors visit the emergency room each year with injuries related to falls. Falls are the primary cause of fractures and injury/death among older adults1. By 2020, the cost of falls is expected to reach almost 44 billion dollars annually. This includes hospital stays, surgery and home health2.
The Berg Balance Scale (BBS) tests static and dynamic balance activities. The BBS consists of 14 separate items graded on a scale of 0-4. A score below 45 indicates balance impairment3,4. To complete the test, the following items are needed: chairs with and without arms, a stopwatch, ruler and a 6-inch step. The test requires 15-20 minutes to complete3.
Intended Population and Validity
The intended audience for the BBS is elderly individuals. This test determines an individual's fall risk3,5,4,6. Berg et al determined that the scale was valid with patients post stroke7. Smith et al performed a study to determine the best clinical tool for an individual post-acute stroke. The researchers compared the BBS and functional reach test. They concluded that the functional reach test demonstrated similar results to the BBS, in that as the subjects' function deteriorated so too did the subjects' balance8. Whitney et al performed a study to determine the validity of the BBS in people with vestibular dysfunction. The researchers found that the Dynamic Gait Index was more appropriate than the BBS for patients in this population9. According to Thorbahn and Newton, the BBS is a valid test to predict falls in elderly people10. Riddle and Stratford published an article illustrating validity indexes and used the BBS as the example. They evaluated the article written by Thorbahn and Newton and a study written by Shumway-Cook et al. Riddle and Stratford describe the BBS to be valid for elderly individuals (65-94 years old) at risk of falling6. Shumway-Cook et al found that an individual with a history of imbalance and a score of 42 on the BBS would have a 91 percent predicted probability of falling. The researchers state that the BBS is the best single predictor of fall status11.
According to Domholdt, reliability is the "extent to which measurements are repeatable"12. Overall, the BBS appears to be a reliable tool used to measure outcomes. Berg et al demonstrated that the inter-rater reliability ICC was .99. Four therapists repeated the tests in order to measure inter-rater ICC and found the reliability to range from .71-.9913. Bogle Thorbahn and Newton reported an inter-rater reliability of .88. This is lower than that reported by Berg et al. After the researchers analyzed the results, they found that three items were disagreed on the most: reaching forward with outstretched arm, tandem stance and unilateral stance. The researchers did not state a reason for the disagreement10. Domholdt states that a relative reliability of .80-1.0 is "very reliable" and .69 and below constitutes "poor reliability"12. Based on this categorization, the BBS is very reliable.
Minimal Detectable Change
Minimal detectable change (MDC) "represents measurement of detectable error." Stevenson states that any score higher than the MDC is attributed to a change in the state of the patient and not attributed to error14. According to Domholdt, the standard error of measurement (SEM) is the standard deviation of many repeated measures on the same subject12. Only one article was found measuring the MDC. Stevenson analyzed the SEM/MDC to compare the magnitude of change required to establish that an actual change had occurred according to the BBS. Stevenson measured four different groups, and they are described as follows: assist, standby assist, independent and all individuals combined. Stevenson evaluated the results at two different confidence intervals (CI)14. Confidence interval is defined as the uncertainty of measurement. For example, a 95 percent CI would lead to a 5 percent uncertainty of the value for the entire population15. Stevenson used CI of 90 and 95 percent. The SEM for the four different groups, all individuals, independent, standby-assist and assist, are given respectively: 2.49, 2.26, 2.15 and 2.13. The MDC ranges from 5.0-6.8 among all of the groups, but the author states that a change of 6 on the BBS is 90 percent confident that an actual change has occurred in the patient's status. It is assumed that the author is speaking about all individuals, since the MDC for all individuals is 5.8. This assumption is made because the MDC for those who require assistance is 6.814.
Sensitivity to Change
How often a diagnostic test detects a disease when it is present is sensitivity, whereas specificity is how often a test is negative when the condition is not present6. Bogle Thorbahn and Newton found the sensitivity of the BBS to be 53 percent, but the specificity was 96 percent. This means that 47 percent of fallers were classified as non-fallers, but only 4 percent of non-fallers were classified as fallers. These numbers mean that the test is suitable to identify non-fallers. Non-fallers were defined as achieving a score of greater than or equal to 45/5610. The authors attributed the low sensitivity to the adoption of strategies to prevent falling by the subjects.
Clinically Significant Difference
Shumway-Cook et al determined the significance of a change in score on the BBS. The authors separated the subjects into two groups: fallers and non-fallers. Fallers were defined as experiencing two or more falls in the past six months, and a fall is any unexpected contact with a supporting surface. The authors found that in the range of 54-56/56, a one-point drop was associated with a 3-4 percent increase in fall risk. They also found that from the range of 46-54/56, a one-point drop was indicative of a 6-8 percent increase in fall risk. Finally, a score below 35/56 led to a 100 percent risk of falling11. Therefore, the individual's risk of falling depends on where in the scale the patient scores.
Use in the Clinic
The BBS may have a ceiling effect, and is insensitive to individuals with very high levels of balance10. Patients that score 40 or less are appropriate for balance and mobility training to reduce fall risk11. The BBS is appropriate for community dwelling elderly patients, patients that have suffered a stroke, and the elderly population with balance difficulties5.
To conclude, the BBS is a test with 14 subsets that evaluates balance. A score of less than 45 is indicative of balance impairment. This test can be administered with little cost and takes 15-20 minutes to complete.
Vincent Gutierrez graduated from Governors State University (IL) in 2007 with a Master of Physical Therapy degree. He currently practices at Physical Therapy and Spine in Tinley Park, IL, which is a MDT based clinic emphasizing spinal care.
1. NIH Senior Health. About Falls. Retrieved from the World Wide Web, www.nihseniorhealth.gov
2. Centers for Disease Control and Prevention. Falls Among Older Adults. Retrieved from the World Wide Web, www.cdc.gov
3. O'Sullivan S.B., Schmitz T.J. Physical Rehabilitation. 5th ed. Philadelphia, PA: F.A. Davis Company; 2007:255-257.
4. Zwick D., Rochelle A., Choksi A., Domowicz J. Evaluation and treatment of balance in the elderly: A review of the efficacy of the Berg Balance Test and Tai Chi Quan. Neuro Rehab. 2000;15:49-56.
5. Steffen T.M., Hacker T.A., Mollinger L. Age- and Gender-Related Test Performance in Community-Dwelling Elderly People: Six-Minute Walk Test, Berg Balance Scale, Timed Up & Go Test and Gait Speeds. Phys Ther. 2002;82(2):128-137.
6. Riddle D.L., Stratford P.W. Interpreting Validity Indexes for Diagnostic Tests: An Illustration Using the Berg Balance Test. Phys Ther. 1999;79(10):939-948.
7. Bergk, Wood-Dauphinee S., Williams J. The Balance Scale: reliability assessment with elderly esidents and patients with an acute stroke. Scand J Rehab Med. 1995;27:27-36.
8. Smith PS, Hembree J.A., Thompson M.E. Berg Balance Scale and Functional Reach: determining the best clinical tool for individuals post acute stroke. Clin Rehab. 2004;18:811-818.
9. Whiney S, Wrisley D., Furman J. Concurrent validity of the Berg Balance Scale and the Dynamic Gait Index in people with vestibular dysfunction. Physio Research Int. 2003;8(4):178-186.
10. Bogle Thorbahn L.D., Newton R.A. Use of the Berg Balance Test to Predict Falls in Elderly Persons. Phys Ther. 1996;76(6):576-585.
11. Shumway-Cook A., Baldwin M, Polissar NL, Gruber W. Predicting the Probability for Falls in Community-Dwelling Older Adults. Phys Ther. 1997;77:812-818.
12. Domholdt E. Rehabilitation Research: Principles and Applications. 3rd ed. St. Louis, MS: Elsevier Inc;2005:258,558.
13. Berg K.O., Wood-Dauphinee S.L., Williams J.I., Gayton D. Measuring balance in the elderly: preliminary development of an instrument. Physio Canada. 1989;41:304-311.
14. Stevenson T.J. Detecting Change in Patients with Stroke Using the Berg Balance Scale. Australian J of Physiother. 2001;47:29-38.
15. Google. Definition of Confidence Interval. Google home page. Retrieved from the World Wide Web, www.google.com