|Year : 2016 | Volume
| Issue : 2 | Page : 42-48
Psychometric properties of the arabic Tinetti Assessment Battery of gait and balance (TAB)
Murad O Al-Momani
Chairman of Audiology and Balance Unit ENT Department, College of Medicine, King Saud University, Riyadh, Saudi Arabia
|Date of Web Publication||6-Jan-2020|
Ph.D Murad O Al-Momani
P O. Box 245 Riyadh 11411
Source of Support: None, Conflict of Interest: None
Objective: To investigate the accuracy of translating and adapting the Tinetti Assessment Battery for gait and balance (TAB) from English into Arabic and to assess the psychometric properties of the Arabic version.
Method: A forward-backward translation process was used to translate the measure. Two hundred and twenty-one nursing home residents participated in this study. Participants were assessed using the translated Arabic and the original English versions of the TAB.
Results: Cronbach alpha for the Arabic version (n = 221) of the TAB was 0.97. Test-retest reliability of the Arabic version (n = 25) ranged from 0.75-1.00 (0.94). The percent agreement for each of the Arabic and English items (n = 25), using Kappa statistics, ranged from 0.75 to 1.0 (total 0.89). Construct validity of the Arabic version (n=221) was tested using factor analysis. Two factors were rotated using a Varimax rotation procedure. The factors’ classification and the items’ distribution related to these factors were very similar to the English version of the TAB. Only 2 items of the 17 items had different distribution than the original version.
Conclusions: The Arabic version of the TAB has very good psychometric properties and can be utilized with confidence with Arabic population.
Keywords: Psychometric properties, Tinetti Assessment Battery (TAB), gait, balance
|How to cite this article:|
Al-Momani MO. Psychometric properties of the arabic Tinetti Assessment Battery of gait and balance (TAB). Saudi J Otorhinolaryngol Head Neck Surg 2016;18:42-8
|How to cite this URL:|
Al-Momani MO. Psychometric properties of the arabic Tinetti Assessment Battery of gait and balance (TAB). Saudi J Otorhinolaryngol Head Neck Surg [serial online] 2016 [cited 2021 May 5];18:42-8. Available from: https://www.sjohns.org/text.asp?2016/18/2/42/275262
| Introduction|| |
During the last century, Rehabilitation therapy has become increasingly prominent in Arabic speaking countries. There are currently several institutions that offer Rehabilitation therapy programs in the Arabic countries (e.g. Jordan, Saudi Arabia, Lebanon, and Kuwait), and nearly all of these educational programs teach in English Language since the scientific literature is written primarily in English. English instruction ensures that the rehabilitation therapy graduates can maintain high standards of practice during their career by accessing the latest advancements in science and evidence based practice. Despite the rehabilitation therapist’s English proficiency there are barriers facing rehabilitation therapists when administering standardized assessments to their Arabic clients. As English is the main language for science and publications, the majority of the rehabilitation therapy standardized assessments are developed and written in English. Only few of these instruments are translated into Arabic language and thus more translated instruments are needed.
Translating standardized assessments from one language to another is a common procedure in the global healthcare community. Examples of standardized assessments that have been translated into Arabic include the Mini Mental State Examination - MMSE , the Montreal Cognitive Assessment – MoCA , Disability of Arm, Shoulder, and Hand Assessment - DASH  among several others. However, translating an instrument from one language to another is complex procedures that must take into account both cultural and language differences ,,,.
The most common procedure for the translation is the backward-forward translation method. During the forward translation, a group of bilingual researchers with expertise in the content area translate the instrument from the original language (e.g. English) to the target language (e.g. Arabic). This is followed by backward translation in which a bilingual individual who is not familiar with the original instrument translates the instrument from the target language (e.g. Arabic) back to the original language (e.g. English). Then, the backward translated items are compared with the original instrument to determine any differences, deletion of words and changes in meaning.
After the instrument is translated, the accuracy of the translation may be tested by administering the instrument to individuals that are proficient in both languages. All individuals complete both language versions with the order of completion counter-balanced and then the two versions are compared. If there are significant differences in responses between the two language versions, this suggests poor translation accuracy. However, interpretation must be made cautiously because some of the differences may occur due to language proficiency issues, cross-cultural context, or test-retest reliability and not necessarily the inaccuracy of the translation ,.
The Tinetti Assessment Battery (TAB)  is the oldest and the most widely used clinical instrument to test gait and balance as well as to predict probability of fall in elderly population . One main advantage of the TAB that it assesses both gait and balance. Other advantages are related to its high inter-rater reliability (85%) as well as excellent sensitivity of around 93% . TAB also exhibits excellent test-retest reliability and good discriminate, convergence and predictive validity . This instrument has been translated to few other languages like Turkish  and Taiwanese  but never to Arabic. Up to our knowledge, there is no instrument in Arabic designed to assess gait and balance deficits.
The main objective of this study was to describe the process of translating and adapting Tinetti Assessment Battery (TAB) into Arabic language. We also investigated the accuracy of the translation and adaptation processes in addition to the psychometric properties of the Arabic translated TAB. We examined the extent to which the Arabic translation of the instrument maintains the same construct of the original version. An accurate and reliable translation will allow rehabilitation therapists to use the instrument with Arabic individuals. As a result rehabilitation therapists will be able to use the TAB to develop treatment plans for addressing gait and balance deficits among Arabic population.
| Materials and Methods|| |
Design: This study used a forward backward translation method to translate and adapt TAB. Additionally; it used a cross sectional, descriptive, and survey methodology to study psychometric properties of TAB.
Jordan has 11 nursing homes. These nursing homes house approximately 290 residents, males and females, with the mean age of 60 years. This study took place in Jordan (January 2012 to February 2013). All nursing home residents in Jordan were invited to participate in this study. Participants signed the consent form before data collection, which was approved by the Institutional Research Board (IRB) at King Abdullah University Hospital (KAUH) and the deanship of scientific research at Jordan University of Science and Technology (JUST). Thirty-three residents (11.4% of the total residents) were excluded from the study because of mental retardation or severe dementia. 36 residents (12.4% of the total residents) refused to participate in the study. Two hundred and twenty-one residents (76.2% of the total residents) with a mean (SD) age of 62.4 (13.9) years agreed to participate in the study.
TAB is a simple, easily administered test (10-15 minutes) that measures an individual’s gait and balance. The test assesses the individual’s ability to perform specific tasks. For example, the individual is asked to rise from a chair, stand on one foot with eyes opened and then eyes closed, walk for 10 steps, etc. Scoring of the TAB is done on a three point ordinal scale with a range of 0 to 2 for each item. A score of 0 indicates the highest level of impairment; while a score 2 indicates the highest level of independence of the individual.
The individual scores are then combined to form three measures; an overall gait assessment score, an overall balance assessment score, and a gait and balance score. The maximum score for the gait component is 12 points. The maximum score for the balance component is 16 points. The maximum total score is 28 points. In general, patients who have a total TAB score below 19 are at high risk for falls and have balance and gait impairments . Concurrent validity ranges from 0.64-0.70. Inter-rater reliability reported from 0.75-0.90. Tests re-test reliability reported from 0.88-0.97 for different studied populations ,,. In this study, this instrument was translated and adapted to be used for Arabic population.
The English version of the TAB was translated into Arabic using a backward-forward translation process . In the forward translation, four bilingual faculty members expert in the process of forward-backward translation (two from Jordan University of Science and Technology/Jordan, one from King Saud University/ Saudi Arabia, and one from Alsharjah University/United Arab Emirates) translated the original instrument from English into Arabic. For those items in which there were differences of opinion regarding the translation, the differences were discussed and a final agreement was reached by consensus. These initial versions were pilot tested with 20 healthy participants who were asked to indicate items that were unclear. Based on their feedback, items were revised and the revised versions were administered to another 20 participants who also were asked to indicate unclear items, which then modified. At this point no further modification of the Arabic version of the translated instrument was needed. The Arabic version of the measure was translated back into English by a bilingual person who was not familiar with the original version of the instrument. This person is a native English speaker who speaks Arabic fluently. Ten evaluators (assistant professors in rehabilitation sciences) determined the extent to which the backward translated instrument items were similar to the original version of the instrument. Similarity of linguistic forms implied that words, phrases, and sentence structure of the backward translated items were similar to the corresponding original items. Similarity of meaning implied that the backward translated items conveyed the same subject matter as the original items even though the wording was not the same. Scores ranged from 0 (not similar) to 1 (similar). A cut score of 0.80 or higher was established to determine the adequacy of the Arabic translation. This cut score implied that at least 80% of the evaluators agreed that the backward translated items were similar to the original items in the instrument. A score below 0.80 suggested a possible problem with the translation. At the end of the translation process, and after modifications of some translations, all translated items reached a minimum cut score of 0.80.
A trained bilingual rehabilitation therapist collected demographic information (age, gender, education level) and medical history from participants. The Arabic version of the TAB was performed for all subjects (n=221) by the same rehabilitation therapist.
Before data collection, the rehabilitation therapist was calibrated by a university faculty member by co- examining 20 patients at King Abdullah Univer-sity Hospital using Arabic TAB prior to study commencement until calibration was confirmed. There was a 92.4% agreement between the examiner and trainer.
The percent agreement and the Kappa values for the English and the Arabic versions of TAB were calculated by administering both versions of the test on 25 subjects. The Arabic version was administered first by the trained bilingual rehabilitation therapist. Two weeks later, the same therapist administered the English version on the same 25 subjects. Results of the Arabic version were kept hidden from the therapist to avoid possible bias. The two-week time period was used to avoid remembering of the Arabic version test results by the therapist, which in turn decreases the bias effect.
Additionally, test-retest intra-rater reliability of the Arabic version was calculated using Kappa statistics on 25 subjects. The trained rehabilitation therapist administered the Arabic version of TAB on 25 subjects. Two weeks later, the same therapist administered the Arabic version again on the same subjects. Results of the first time testing were kept hidden from the therapist to decrease chance of bias. The two-week time period helped to keep the first test results hard to remember by the therapist.
Data analysis and management
Data were entered into a personal computer and analyzed using the Statistical Package for Social Sciences (SPSS) software version 11.0 (SPSS®: Inc., Chicago, IL, USA). Cronbach’s alpha coefficients were calculated as a measure for internal consistency for the translated instrument. To determine construct validity, factor analysis using principal component factor analysis was performed, with a factor loading above 0.35 considered acceptable to determine the factor structure of the translated items of the TAB. The percent agreement for each of the Arabic and English items was tested using Kappa statistics. In addition, test-retest reliability of the Arabic version repeated on same subjects was studied using Kappa statistics. The level of significance was set at p ≤ 0.05.
| Results|| |
Two hundred and twenty-one residents were included in the current study. These residents ranged in age between 28 and 100 years with a mean (SD) of 62.4 (13.9) years. Slightly less than one half (45.2 %) of them were females. About one fifth (21.7 %) of subjects had an educational level higher than secondary education, [Table 1].
|Table 1: socio-demographic, personal characteristics, and frequency distributions of medical conditions of study sample (N=221)|
Click here to view
[Table 1] also shows the medical conditions and diseases among participants. About two thirds (60.7%) of the sample had at least two chronic diseases; the most prevalent diseases found among them were psychiatric disorders (45.7%), followed by hypertension and diabetes mellitus, at 33.5% and 23.5% respectively.
In evaluating the linguistic form of the items of the translated instrument, all items ranged between 0.80 - 1.00. This indicates that at least 8 of the 10 reviewers found that the backward translated items adequately matched the original items in terms of phrases, words and structure for all items of the instrument. These results indicate that the translation resulted in an accurate representation of the TAB.
The internal consistency estimates of reliability were computed for the translated TAB. The Cronbach’s alpha coefficients values for the Arabic version of the TAB = 0.97. [Table 2] indicates that the Arabic translated instrument had very good reliability and the internal consistency of the Arabic version was comparable to the reported internal consistency of the English version of 0.91 .
The principle components factor analysis performed on all items of the TAB revealed two factors. The items loading on each factor was very similar to the factors presented in the original instrument, except for 2 items (out of the 17 items). The item of “Turning 360” of factor 1 and the item of “Walking stance” of factor 2 were different than in the original English TAB. Tests of the suitability of the data for the factor analysis proved satisfactory, with the Laiser Meyer Olkin test value being 0.60, which is statistically significant (p<0.05). The cumulative percentage of variance accounted for by factor 1 is 41.405% and for factor 2 is 39.951%, [Table 3].
|Table 3: Factor Analysis Results (Rotated Component Matrixa ) for Arabic version of TAB. Extraction Method: Principal Component Analysis Rotation Method: Varimax with Kaiser Normalization. (n = 221)|
Click here to view
Cronbach’s alpha coefficients values for Arabic and English versions were 0.90 and 0.89 respectively (n = 25). This indicates that the Arabic translated items have very good reliability and that the internal consistency of the Arabic version is comparable to the internal consistency of the English version, p ≤ 0.05. The percent agreement and Kappa values for the items of the English and the Arabic versions of the TAB and the test- retest reliability of the Arabic TAB are presented in [Table 4]. The percent agreement for all items ranges from 0.75 to 1.00 (Arabic-English) and from 0.76 to 1.00 (Arabic-Arabic).
|Table 4: Percentage of agreement on item similarity of the back translation and overall kappa agreement values for Arabic and English versions of TAB|
Click here to view
| Discussion|| |
The reliability of the Arabic TAB was supported by the results of this study. The similarity of the Arabic version to the English version in terms of the meaning of items, internal consistency, test-retest reliability, and construct validity suggest that the Arabic adaptation of the TAB can be used reliably. Internal consistency provides an estimate of consistency among different items, which measures whether different items that measure the same general construct produce similar scores. The internal consistency values found for gait or balance separately and the overall score for gait and balance together of the Arabic version of the TAB were very similar to the original English version. The overall coefficient alpha for the Arabic version was 0.97 and for the English version was 0.91 , indicating that the Arabic translated items have an overall very good internal consistency reliability. This suggests that the Arabic translated items generated equally consistent responses. Similarly, the Turkish version of the TAB, which was tested on 80 subjects (65 years or older) found a Cronbach’s alpha of 0.88 and a good test re-test reliability .
In order to accurately test the validity of an instrument like TAB, it requires comparing the results of both instruments that test the same construct (e.g. gait and balance deficits) . Up to our knowledge, there is no Arabic instrument available to test gait and balance to compare with. Therefore, principle component factor analysis was used to test construct validity of the Arabic version of TAB.
The 17 items of the Arabic version of the TAB were analyzed using principle component factor analysis. Around 88% of the items (15 items) were correctly related to the same factors as the original English version. This result indicates very good construct validity of the Arabic version of TAB. Similarly, Faber et al. (2006) reported good to excellent validity of TAB tested on 245 nursing home residents .
On the other hand, two items in the Arabic TAB did not relate to the same factors as in the original English version; “Walking stance” and “Turning 360”. This is not surprising given that there are similar problems with distinguishing these two items in the English version. Both of these two items could relate to gait and balance factors simultaneously. For example, “Walking stance” requires good balance to stand up and good gait ability to step forward. Similarly, “Turning 360” requires steady posture, which needs good balance. Also, it requires continuous steps that needs normal gait. Overall the factor structure of the Arabic TAB is very similar to the original English version.
The linguistic evaluation of the 17 items shows that the ten reviewers agreed that 80% of the back translated items adequately matched the original items in terms of phrases, words and structure. This was further supported by very good percent of agreement found between the items of the Arabic and English versions (Kappa range 0.75-1.00) and strong test re-test reliability of the Arabic version (Kappa range 0.76-1.00), [Table 4]. Similarly, Kloos et al. (2004) reported similar intra- rater reliability of TAB (Kappa range: 0.62-1.00) in a sample of 21 subjects . Also, Cipriany-Dacko et al. (1997) found good to excellent reliability of TAB on 24 participants aged 60-92 years old .
On the other hand, there were some items in this study with relatively weaker levels of agreement, which could be due to the differences in the administration procedure of some items, variability in scoring of these items, or changing in participants’ status between testing periods. For example, the lowest test re-test agreement (Kappa = 0.75) was for item “Nudged”, which means to push the participant gently. First, it is difficult to calibrate the amount of pushing against the participant performed by the examiner across testing trials. Second, this item has 3 point scoring (0, 1, or 2), not like most other items that only have 2 point scoring (0 or 1) with clearer cut points. The three point scoring may increase the chance of variability in scoring across testing trials. Third, participants’ general health may change in the two week period (between the first and second testing trial), which may affect their gait and balance status. Similarly, relatively weaker agreements were obtained for “Path” and “Trunk’ items. Both items have 3 point scoring defining magnitude of deviation in the path and sway in the trunk. For example, it is difficult for the examiner to differentiate between mild, moderate, and marked deviation in path or sway.
Finally, this study has some potential limitations. First, the study sample has a wide age range and nearly two thirds of the study sample is over the age of 55. This could affect the results of this study because gait and balance deficits are more prone to change from one time to another in elderly population, which might affect reliability measures. Second, our sample consisted of subjects with different medical problems, however, homogeneous sample is recommended in this type of study. Third, since only one examiner was calibrated to test all participants, inter-rater reliability was not measured in this study.
| Conclusion|| |
The overall internal consistency, test re-test reliability measures, and construct validity of the Arabic TAB appear to capture the essence of the original English TAB. These findings provide additional support for the psychometric properties of the original TAB. Therefore, the translated Arabic TAB can be used with confidence to assess gait and balance deficits in Arabic population.
Author of this study would like to acknowledge and thank the nursing homes’ managers for their help and support. The author also extends sincere appreciation to colleagues who participated in the verification of the translated material or data collection: Fidaa Almomani, Ph.D., Wejdan Bani Issa, Ph.D., Wafaa Alqudah, M.S., Ghadeer Ashour, OTR.
| References|| |
Al-Rajeh S, Faan M, Ogunniyi A, Awada A, Daif A, Zaidan R. Preliminary Assessment of an Arabic version of the Mini-Mental State Examination. Ann Saudi Med.
Abdel Rahman T, Gaafary M Montreal Cognitive Assessment Arabic Version: Reliability and validity prevalence of mild cognitive impairments among elderly attending geriatric clubs in Cairo. Geriatr Gerontol Int.
Alotaibi M. Cross-cultural adaptation process and pilot testing of the Arabic version of the Disability of the Arm, Shoulder and Hand (DASH-Arabic). Hand Ther.
Butcher JN. Translation and adaptation of MMPI-2 for international use. In : International adaptation of the MMPI- 2: Research and clinical applications. J.N. Butcher, (Eds). University of Minnesota Press, Minnesota,
pp. 26-43, 1996.
Hachey R, Jumoorty J, Mercier C. Methodology for validating the translation of test measurements applied to occupational therapy. Occup Ther Int.
Ben-Porath YS, Almagor M., Hoffman-Chemi A, Tellegen A. A cross-cultural study of personality with the Multidimensional Personality Questionnaire. J Cross Cult Psycho.
Geisinger KF. Cross-cultural normative assessment: Translation and adaptation issues influencing the normative interpretation of assessment instruments. Psychol Assess.
Neuman A, Greenberg D, Labovitz D, Suzuki L. Cross-cultural adaptation of the sensory profile: Establishing linguistic equivalency of the Hebrew version. Occup Ther. Int
Brislin RW. Translation and content analysis of oral and written materials. In : Handbook of cross cultural psychology. J.W. Barry (Eds). Allyn & Bacon, Boston,
pp. 389-444, 1980.
Tinetti ME. Performance oriented assessment of mobility problems in elderly patients. J Am Geriatr Soc.
Yelnik A, Bonan I. Clinical tools for assessing balance disorders. Neurophysiol Clin.
Maki BE, Holliday PJ, Topper AK. A prospective study of postural balance and risk of falling in an ambulatory and independent elderly population. J. Gerontol
Lin MR, Hwang HF, Hu MH, Wu HD, Wang YW, Huang FC. Psychometric comparisons of the timed up and go, one-leg stand, functional reach, and Tinetti balance measures in community-dwelling older people. J Am Geriatr Soc.
Yücel SD, Şahin F, Doğu B, Şahin T, Kuran B, Gürsakal S. Reliability and validity of the Turkish version of the Performance-Oriented Mobility Assessment. Eur Rev Aging Phys Act.
Faber MJ, Bosscher RJ, Van Wieringen PC. Clinimetric properties of the Performance-Ori en ted Mobility Assessment. Phys Ther.
Kloos AD, Bello-Haas VD, Thome R, Cassidy J, Lewis L, Cusma T, Mitsumoto H. Interrater and Intrarater Reliability of the Tinetti Balance Test for Individuals with Amyotrophic Lateral Sclerosis. J Neurol Phys Ther
Cipriany-Dacko LM, Innerst MD, Johannsen J, Rude V. Interrater Reliability of the Tinetti Balance Scores in Novice and Experienced Physical Therapy Clinicians. Arch Phys Med Rehabil.
[Table 1], [Table 2], [Table 3], [Table 4]