|Year : 2022 | Volume
| Issue : 1 | Page : 22-29
Cross-cultural adaptation and reliability of the berg balance scale in Hindi
Shweta Ravindra Shah1, Vasanti Joshi2, Nicky J Snowdon3
1 Hounslow and Richmond Community Healthcare NHS Trust, Sancheti College of Physiotherapy, Pune, Maharashtra, India
2 Consultant Neuro Physiotherapist, Sancheti College of Physiotherapy, Pune, Maharashtra, India
3 Department of Allied Health Professions, College of Health, Wellbeing and Life Sciences, Sheffield Hallam University, Sheffield, UK
|Date of Submission||13-Oct-2021|
|Date of Decision||11-May-2022|
|Date of Acceptance||07-Jun-2022|
|Date of Web Publication||30-Jul-2022|
Mrs. Shweta Ravindra Shah
Heart of Hounslow Centre for Health, 92 Bath Road, Hounslow, TW3 3EL
Source of Support: None, Conflict of Interest: None
Context: Use of outcome measurement is an important part of modern professional practice. The Berg Balance Scale (BBS) appears to be used in many Indian rehabilitation centres, but no official version has been previously available in Hindi. The BBS was translated into Hindi by the current authors according to accepted guidelines and tested for reliability.
Aims: To translate the Berg Balance Scale into Hindi and investigate the inter-rater reliability when applied to geriatric patients.
Setting/Design: Translation of BBS in Hindi according to guidelines for cross-cultural validation of an outcome measurement was conducted at a renowned institute for orthopedic and rehabilitation in India.
Methods and Materials: The study was approved by the University Research Ethics Committee. The study involved an initial cross-cultural translation following four stages: 1) Forward Translation 2) Synthesis of translation 3) Backward Translation 4) Expert Committee. Expert committee was formed by purposive sampling comprising of experienced health professionals and language experts. For reliability testing, thirty patient participants were recruited from an Indian Orthopaedic hospital. Reliability testing was done by two experienced physiotherapists.
Statistical Analysis Used: Weighted Kappa (Kw) was used to assess inter-rater reliability for the total scores and each individual item of BBS. Intraclass correlation coefficient was calculated to make comparisons of the inter-rater reliability with results of other studies.
Results: The Intra-class correlation coefficient (ICC) for the total score of the BBS was 0.977(95% Confidence Interval= 0.951-0.989). The Kw for total score of BBS was 0.819 (95% Confidence Interval = 0.734-0.905).
Conclusion: Hindi version of BBS appears to have excellent inter-rater reliability when used by experienced physiotherapists for rehabilitation of geriatric patients.
Keywords: Berg Balance Scale, Cross-cultural adaptation, Reliability, Validity
|How to cite this article:|
Shah SR, Joshi V, Snowdon NJ. Cross-cultural adaptation and reliability of the berg balance scale in Hindi. Indian J Phys Ther Res 2022;4:22-9
|How to cite this URL:|
Shah SR, Joshi V, Snowdon NJ. Cross-cultural adaptation and reliability of the berg balance scale in Hindi. Indian J Phys Ther Res [serial online] 2022 [cited 2022 Nov 27];4:22-9. Available from: https://www.ijptr.org/text.asp?2022/4/1/22/353019
| Introduction|| |
Postural instability is a serious problem in the elderly and causes significant morbidity and mortality. One-third of those over 65 years of age fall one or more times every year, and it increases to 40% for those over 80 years of age., Balance impairment is said to be a key risk factor for falls in the elderly with a relative risk of 2.9. Fall in the elderly is an important public health issue and can be prevented by identification and treatment of the modifiable risk factors. The intrinsic risk factors for falls include vertigo, sedation, neurological disorders leading to balance impairment, cognitive dysfunction, visual problems, musculoskeletal abnormalities, and postural instability. The extrinsic risk factors include insufficient environment lighting, use of assistive devices or walking aids, and improper footwear., However, balance and postural control remain the key risk factor because the ability to maintain balance and posture is required to carry out all daily tasks ranging from standing and walking to sitting and getting up from the chair, and hence, enhancement of balance should be the main objective to prevent falls in geriatric population.
The Berg Balance Scale (BBS) is an instrument for assessing functional balance. It measures the postural control domain required in daily life, like the ability to respond to voluntary movements of the body and reaction to external perturbations. It is a clinical outcome measure developed to assess frail elderly individuals with balance deficits regardless of their age. It is the most common measurement tool used in rehabilitation amongst different clinical and laboratory measures available for balance assessment. With research projects adapting a multicultural perspective, it is vital to adapt the health status measures in different languages other than the source language., Most of the health status measures have been developed in English and this restricts their use only to the English-speaking population. The BBS has been translated and validated in various languages such as Brazilian, Italian, Turkish and Norwegian.,,, Reliability studies on the scale in each of the language show good inter and intra observer reliability.,,, In the authors' experience, this scale is in common use in India. However, the lack of an accepted Hindi language version must mean that it is translated by individual practitioners, which is likely to introduce error.
Hindi being the national language of the country is understood by most patients and adapting the Berg Balance Scale to Hindi will be of great benefit to patients and therapists and will be a step forward in standardisation of outcome measures. Psychometric evaluation of a translated scale is vital to ensure that it retains good measurement properties.
Reliability is a statistical measure of how reproducible the instrument's data are and how consistently or dependably does an instrument measure what it is supposed to be measuring. During rehabilitation, a patient may be assessed by more than one physiotherapist and hence it is essential to have high inter- rater reliability.
The aim of this study was to translate the BBS into Hindi by following the approach for cross cultural adaptation of instruments and to assess the inter-rater reliability in an Indian population.
| Subjects and Methods|| |
This was a cross-cultural adaptation study including reliability testing. The study was reviewed and approved by the university research ethics committee.
Informed written consent for participating in the research was acquired from patient participants.
BBS is a 14-item test that takes into consideration common daily life tasks. The maximum possible score is 56 and each item further has an ordinal scale of 5 alternatives ranging from 0(unable to perform) to 4(normal performance). It is a simple, easy, and safe test for the evaluation of balance in elderly people., A cut off score of 45 is the best model for prediction of falls. It takes only 15 minutes to administer it and requires simple equipment like a watch and ruler. It is found to have good construct, concurrent and predictive validity.] It has excellent inter and intra observer reliability which is 0.98 and 0.99 respectively.The internal consistency is also high (Cronbach's alpha = 0.96) which shows that the scale measures one concept and provides information on balance.
The translation was carried out by following the guidelines of Beaton et al (2000) and Wild et al (2005), which is an accepted method for translating a measure.
An expert committee was formed by purposive sampling by the main author which comprised of experienced health professionals and language experts. The participants involved in every stage of translation are described in [Table 1]
In this stage, two translations were done in Hindi: one by a registered translator and the other by two students of MSc neurological physiotherapy who were fluent both in Hindi and English. Since the second translator was unaware of the clinical perspective, he used a language that is more commonly used by the general population and hence was beneficial.
Synthesis of translation/reconciliation
The two translated versions were discussed by the panel members. Discussions were carried out between experienced professionals at a renowned institute for orthopedics and rehabilitation and were communicated to the student members at the UK University via email. They came to a common consensus after resolving the discrepancies.
This was done by a registered translator not involved in the initial two stages. It was also translated by a second translator who is fluent in both languages and then both the translated versions were compared. The discrepancies in the original and translated version were resolved and synthesized to one prefinal version by the expert committee.
This stage was essential for validity checking so that it can be assured that the translated version reflects the same content as original version.
This involved input from all the members in the above stages to have cross-cultural equivalence. The main role of this committee was to consolidate all the versions of the scale and develop a prefinal version for testing.
Procedure for reliability testing
Reliability testing was done by two experienced physiotherapists who had used BBS previously in their clinical practice.
The patients were tested only on one occasion by both the raters as there would have been difference in conditions if they were tested separately by both raters. One rater gave the commands and both observed and gave independent ratings for each patient participant. Raters decided themselves who would give commands and interchanged from one patient to the next.
Participants for reliability testing
Thirty patient participants older than 65 years of age were recruited from an Orthopaedic and Rehabilitation Hospital in India. The participants were recruited from the physiotherapy outpatient department of the hospital and the spine and knee club, where group exercise classes are taken for patients with mechanical back and knee pain. This recruitment strategy was driven by convenience. Participants did not need to have current problems with pain to be involved in the study.
Patients above 65 years of age who were able to walk with or without a walking aid, and could understand and speak Hindi, were included in the study. The exclusion criteria were patients with recent fracture as their performance could been hampered by pain, bed-bound or wheelchair- dependant patients, patients having amputation of lower extremities and patients having cognitive deficits.
The mean age of the participants was 68.86 years (range 65-81), 18 were women and 12 men. Only three subjects used a cane for assistance, whereas others did not use any walking aid. Six subjects reported fall in the past six months out of which five had only one fall and one subject had two falls. All the demographic data were collected from the patient participants after informed consent [Table 2].
Normality test was done on the data and it was found that the data were not distributed normally (P < 0.013) and hence it was considered ordinal data.
Therefore, Weighted Kappa (Kw) was used to assess inter-rater reliability for the total scores and each individual item of BBS. This is done to distinguish serious disagreement between the examiners and to find out the weights of their disagreement. A kappa value of >0.75 indicates excellent agreement, 0.4-0.75 is fair to good agreement and <0.4 is suggested as poor agreement. Weighted Kappa was calculated with Medcalc. MedCalc Software Ltd, Belgium.
Intraclass correlation coefficient was calculated using SPSS 19 (IBM, Armonk, NY, USA) to make comparisons of the inter rater reliability with results of other studies. An ICC of 0.80 or above reflects high reliability, 0.60 to 0.79 is moderate reliability and less than 0.60 is indicated as poor reliability.,
95% confidence intervals were used as they provide the boundaries within which the sample means lie to 95% confidence. If the interval is wide, it indicates variability of the measure or sample but if the interval is small, it would mean we can have more confidence in the results.
| Results|| |
Translated Hindi version of Berg Balance Scale
Following the process of cross-cultural adaptation and expert committee agreement, a Hindi version of Berg Balance Scale was produced [Table 3].
There were only minor differences among the expert committee reviewers and the translators in the stages of initial and back translation. The exact translation of Item 3 (sitting with back unsupported, but feet supported on a floor a stool) in Hindi depicted a different meaning and hence it was modified after discussion. Similarly, there was no difference in the instructions for ratings 4 and 3 of Item 13 after translation and hence they were modified after discussion. There was confusion in the use of Hindi words for the words “support” and “assistance” and they were used interchangeably for different items. It was therefore finally decided to use the word “सहारा” for support and the word “सहायता” for assistance.
Reliability testing results
A total of 30 patient participants (12 males and 18 females) were included. The median of age was 67, interquartile range (IQR) 7, range 16 (81–65), and mean was 68.867 (± 4.329).
The mean value of total score of BBS scored by Rater 1 was 48.86 (± 5.618). The median was 51, IQR 5, and range 26 (55–29).
The mean value of total score of BBS scored by Rater 2 was 48.50 (± 5.952). The median was 49.5, IQR 5, and range 29 (55–26).
The raters had a good agreement between scores for each individual item of the scale. The kw value for all the 14 items ranged from 0.592 to 1.000 with a mean value of 0.760 [Table 4].
|Table 4: Distribution of scores, weighted kappa, and 95% confidence interval for each item from one of the raters|
Click here to view
There was 100% agreement (kw=1.000) for items 6 to 9 and 12; whereas item 4 (standing to sitting, kw=0.627) and item 5 (transfers, kw=0.592) showed fair agreement. There was a lot of difference between scores of Rater 1 and Rater 2 for item 10(turning to look behind).
The weighted kappa for the total score was 0.819(95% confidence interval= 0.734-0.905 and the ICC for the total score between the 2 Raters was 0.977 (95% confidence interval= 0.951-0.989).
| Discussion|| |
The aim of this study was to test the instrument for cultural equivalence to be applied to the Indian population and evaluate the reliability of this Hindi version of BBS.
The scale was translated without any variations from the original scale as the initial focus of the investigator was to translate the scale to Hindi relatively literally, keeping in mind the metropolitan population of Indians.
The investigators used two therapists who were already experienced in use of BBS as the focus was to test the reliability of the translated version and it was thought that therapists who already know the scale were better suited as it negates any initial training required for using the scale.
The Hindi version of BBS was found to have excellent inter-rater reliability. Inter-rater reliability was shown to be similar to that reported for other versions of the BBS,,,, [Table 5] gives a brief overview of the ICC value for reliability of various versions of BBS.
|Table 5: Intra-class correlation coefficient values for inter-rater reliability for various versions of Berg Balance Scale|
Click here to view
The weighted kappa (kw) values for each item in the present study were between 0.59 to 1.00. Only item 4 (Standing to sitting) & 5 (Transfers) had kw value less than 0.75 suggesting that the items were less reliable, but all other had kw value above 0.75 and hence it suggests excellent inter-rater reliability. This is consistent with the findings from other studies., The variability in scoring between the raters for items 4 & 5 could be due to a difference in understanding of the instructions for the scoring. They have wordings like minor/definite use of hands which is more subjective. However, this appears to be an issue with underlying Berg Balance Scale rather than the Hindi version.
In Item 10 (turning to look behind), there was a lot of difference in the scoring between the two raters. This was consistent with the findings of Halsaa et al. where they had a maximum difference of 2 points. The difference in scoring for item 10 could be due to the instructions for rating of the item. The raters must judge the weight shift when the patient looks behind and this aspect tends to be very subjective as one rater might not perceive the weight shift exactly as the other one.
The weighted kappa for the total score is less than the ICC value. The possible explanation for this could be that since ICC considers the data to be normally distributed the results tend to be inflated. When weighted kappa is used the data is analysed more cautiously, in keeping with the non-normal distribution. ICC has been used here to enable comparison with other studies who have used ICC without checking data distribution first. It may be that those other studies also have non-normally distributed data, in which case the reliability coefficient for the BBS in previous research could be incorrectly inflated.
Strengths and limitations
The approach to translation was robust and the researcher carried out a reliability analysis of both the total scores and individual items.
The rating 3 or 4 were used most often by both the examiners while using the Hindi version. This is consistent with the reports of BBS by Sahin et al, Halsaa et al, and Kornetti et al.,, This reflects that the subjects recruited for the study had higher functional status as they were community dwelling individuals and not patients referred specifically for rehabilitation. The mean value of BBS was higher (48.03) compared to cut off value of 45. It was also higher than the mean value reported by Halsaa et al and Ottonello et al respectively., It suggests low risk of falls in the subjects recruited for the study. All the participants recruited in this study had BBS scores on the higher side and hence good balance as discussed balance. This limits the generalisability of the results only to patients with good balance and might not be applicable to patients with poor balance
The investigators used two therapists who were already experienced in use of BBS. There is a possibility that this might have had a subjective influence on the reliability as they already knew how to rate through their experience using the English version.
Activities like squatting, getting up from a low bed require a high level of balance and function and these form a part of day-to-day activities for a wider Indian population. The investigator acknowledges that not including the item squatting as a part of cross-cultural validation study is a limitation of this study. However, as the focus was on urban Indian population, the use of the translated version in the wider population was not considered initially and could be a scope for further research.
This study presents the Hindi version of BBS obtained by following a validated procedure for cross-cultural translation and the reliability testing shows excellent inter- rater reliability. It can therefore be used as a reliable balance measure in clinical practice and research.
| Conclusion and Scope for Future Research|| |
The Hindi version of BBS appears to have excellent interrater reliability when used by experienced physiotherapists for rehabilitation of geriatric patients. It should enable easier use of BBS in Indian environments because therapists do not need to translate on the spot and can have confidence in the version they are using.
Future research investigating the content validity of this scale in a more diverse Indian population would be valuable. This would enable us to determine whether items such as squatting and getting up from a low mattress should be included in the Hindi version of the BBS. It may be valuable to conduct a further review into the Hindi language in this BBS version, to ensure that it is meaningful and acceptable to a wide range of Hindi speakers.
I would like to thank Sancheti Institute for Orthopaedic and Rehabilitation for permitting the author to do data collection with the help of their patients.
I would like to thank all the participants and members of the expert committee without whom this study would not have been possible.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med 1988; 319:1701-7.
Lajoie V, Giallagher SP. 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. Arch. Gerontol. Geriatr 2004;38:11-26.
Muir S, Berg K, Chesworth B and Speechly M. Use of the Berg Balance Scale for Predicting Multiple Falls in Community Dwelling Elderly People: A prospective Study. Phys.Ther. 2008; 90.
Gillespie LD, Gillespie WJ, Robertson MC. Interventions for preventing falls in elderly people. Cochrane database Syst.Rev. 2003; 4.
Miyamoto S T, Lombardi Junior I, Berg K O, Ramos L R and Natour J. Brazilian version of the Berg Balance Scale. Braz. J. Med. Biol. Res 2004; 37:1441-21.
Ottonello M, Ferriero G, Benevolo E, Sessarego P, Dughi D. Psychometric evaluation of the Italian version of the Berg balance scale in rehabilitation inpatients. Eur J Phys Rehabil Med 2003;39:181-9.
Halsaa K, Brovold T, Graver V, Sandvik L, Bergland A. Assessments of interrater reliability & Internal Consistency of the Norwegian Version of the Berg Balance Scale. Arch Phys Med Rehabil 2007;88.
Anderson RT, Aaronson N, Wilkin D. Critical review of the international assessments of health-related quality of life generic instruments. Qual Life Res1993;2:369-95.
Bullinger M, Alonso J, Apolone G. Translating health status questionnaires and evaluating their quality: the IQOLA Project approach. International Quality of Life Assessment. J. Clin. Epidemiol 1998;51:913-23.
Guillemin F, Bombardier C, Beaton D. Cross-cultural adaptation of health-related quality of life measures: literature review and proposed guidelines. J. Clin. Epidemiol 1993;46:1417-32.
Sahin F, Yilmaz F, Ozmaden A, Kotevoglu N, Sahin T, Kuran B. Reliability & Validity of the Turkish Version of the Berg Balance Scale. J Geriatr Phys Ther; 31.
Bannigan K and Watson R. Reliability and Validity in a nutshell. J Clin.Nurs 2009;18:3237-43.
Berg KO, Maki BE, Williams JI, Holliday PJ, Wood-Dauphinee SL. Clinical and laboratory measures of postural balance in an elderly population. Arch Phys Med Rehabil 1992;73:1073-80.
Berg KO, Wood-Dauphinee SL, Williams JI, Gayton D. Measuring balance in the elderly: preliminary development of an instrument. Physiother Can 1989;41:304-11.
Berg KO, Wood-Dauphinee SL, Williams JI. The balance scale: reliability assessment with elderly residents and patients with acute stroke. Scand. J. Rehabil. Med 1995;27:27-36.
Beaton, Bombraider C, Guillemin F, Ferraz Bosi M. Guidelines for the Process of Cross-cultural Adaptation of Self Report Measures. Spine J 2000;25:3186-91.
Wild D. Principles of good practice for the translation and cultural adaptation process for patient-reported outcomes (PRO) measures: Report of the ISPOR task force for translation and cultural adaptation. Value health 2005;8:94-104.
Cohen J. Statistical power analysis for the behavioural sciences. 2nd
ed. Hillsdale, New Jersey. 1998.
Holey SM, Osberg JS. Kappa coefficient calculation using multiple ratings per subject. A special communication. Phys. Ther. 1989; 69:970-4.
Fleiss JL. The design and analysis of clinical experiments. New York: John Wiley & Sons.1986. p 1-32.
McCluggage WG, Bharucha H, Caughley LM, Date A, Hamilton PW, Thornton CM, et al
. Interobserver variation in the reporting of cervical colposcopic biopsy specimens: comparison of grading systems. J Clin Pathol 1996;49:833-5.
Bland M. An introduction to medical statistics, 3rd
ed. Oxford University press, New York. 2000.
Kornetti DL, Fritz SL, Chiu YP, Light KE, Velozo CA. Rating scale analysis of the Berg Balance Scale. Arch Phys Med Rehabil 2004; 85:1128-35.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]