Cross-cultural adaptation and analysis of the psychometric properties of the Balance Evaluation Systems Test and MiniBESTest in the elderly and individuals with Parkinson's disease: application of the Rasch model

Abstracts

BACKGROUND:

Older adults and individuals with neurological problems such as Parkinson's disease (PD) exhibit balance deficits that might impair their mobility and independence. The assessment of balance must be useful in identifying the presence of instability and orient interventions.

OBJECTIVE:

To translate and perform a cross-cultural adaptation of the Balance Evaluation Systems Test (BESTest) and MiniBESTest to Brazilian Portuguese and analyze its psychometric properties.

METHOD:

The tests were translated and adapted to Portuguese according to a standard method and then subjected to a test-retest reliability assessment (10 older adults; 10 individuals with PD). The psychometric properties were assessed by the Rasch model (35 older adults; 35 individuals with PD).

RESULTS:

The reliability coefficient of the tests relative to the items and subjects varied from 0.91 and 0.98, which is indicative of the stability and reproducibility of the measures. In the BESTest, the person (4.19) and item (5.36) separation index established six balance ability levels and seven levels of difficulty, respectively. In the MiniBESTest, the person (3.16) and item (6.41) separation index established four balance ability levels and nine levels of difficulty, respectively. Two items in the BESTest did not fit with the model expectations, but the construct validity was not compromised. No item in the MiniBESTest was erratic.

CONCLUSIONS:

The results corroborate the diagnostic and screening functions of the BESTest and MiniBESTest, respectively, and indicate that the Brazilian versions exhibit adequate reliability, construct validity, response stability, and capacity to distinguish among various balance ability levels in older adults and individuals with PD.

postural balance; physical therapy; Parkinson's disease; older adults; cross-cultural adaptation; Rasch analysis

CONTEXTUALIZAÇÃO:

Idosos e indivíduos com problemas neurológicos, como a doença de Parkinson (DP), apresentam déficits de equilíbrio que podem comprometer a mobilidade e independência. A avaliação do equilíbrio deve identificar a instabilidade e direcionar a intervenção.

OBJETIVOS:

Traduzir e adaptar transculturalmente o Balance Evaluation Systems Test (BESTest) e o MiniBESTest para o português-Brasil e analisar suas propriedades psicométricas.

MÉTODO:

Os testes foram traduzidos e adaptados para o português seguindo instruções padronizadas e foram submetidos à análise de confiabilidade teste-reteste (dez idosos; dez indivíduos com DP). As propriedades psicométricas foram avaliadas pela análise Rasch (35 idosos; 35 indivíduos com DP).

RESULTADOS:

Os coeficientes de confiabilidade dos testes para itens e indivíduos variaram de 0,91 a 0,98, indicando estabilidade e possibilidade de reprodução das medidas em aplicações subsequentes. No BESTest, os índices de separação dos indivíduos (4,19) e dos itens (5,36) determinaram, respectivamente, seis níveis de habilidade de equilíbrio e sete níveis de dificuldade. No MiniBESTest, os índices dividiram os indivíduos (3,16) em quatro níveis de habilidade e os itens (6,41) em nove níveis de dificuldade. Dois itens do BESTest não se adequaram às expectativas do modelo, o que não comprometeu sua validade de constructo. No MiniBESTest não houve item errático.

CONCLUSÕES:

Os resultados dão suporte à função de diagnóstico e triagem para o BESTest e MiniBESTest, respectivamente, e sugerem que as versões brasileiras apresentam adequada confiabilidade, validade de constructo, estabilidade das respostas e capacidade de discriminação entre diferentes níveis de habilidade do equilíbrio em idosos e indivíduos com DP.

equilíbrio postural; fisioterapia; doença de Parkinson; idosos; adaptação transcultural; análise Rasch

Introduction

Posture control (PC) is the ability to maintain the projection of the body center-of-mass within the limits of the base of support under static and dynamic conditions and involves the control of the body position in space for the purpose of ensuring stability and orientation¹. PC depends on the visual, vestibular, somatosensory, nervous, biomechanical, and cognitive systems^{1 2} that are exposed to different types of dysfunctions.

Older adults are liable to acquire balance disorders as a function of multiple impairments, such as multisensory loss, muscle weakness, and orthopedic and cognitive limitations³. Such impairments cause functional losses, and the balance disorders must be appropriately assessed and treated due to their impact on the quality of life and their high financial onus relative to their health care costs^{4 5}. In addition, several neurological conditions are associated with postural instability, with Parkinson's disease (PD) being the most common movement disorder. The estimated prevalence of PD is 100-200 cases/100,000 inhabitants⁶. From the clinical point of view, PD is characterized by motor dysfunctions such as bradykinesia, tremor, and rigidity. Postural instability tends to appear during the course of the disease and is associated with falls, limited mobility and independence⁷.

Balance disorders are some of the most common problems treated by physical therapists. These health professionals need to identify the patients' instability and select the most appropriate approach to rehabilitation⁸. The current clinical tests for balance assessment focus on the detection of balance disorders and the prediction of the risk of falls^{9 13}. The Berg Balance Scale is one of the most widely used tests. Its ceiling effect has been identified in the evaluation of more subtle balance disorders that occur in older adults or at the initial stages of PD^{12 14}.

Horak et al.⁸ formulated the Balance Evaluation Systems Test (BESTest) to help in the identification of the particular system that may be responsible for poor balance and thus guide the treatment. The BESTest is a versatile instrument that might be applied to individuals at any age and with a broad scope of diseases and severity levels⁸. It comprises 27 items and a total of 36 tasks. The items are clustered in sections corresponding to the six underlying systems that may constrain balance: biomechanical, stability limits, postural responses, anticipatory postural adjustments, sensory orientation, and dynamic balance during gait. Each item is scored on an ordinal four-point scale varying from zero to three (best performance). A shorter version known as the MiniBESTest¹⁴ comprises 14 of the items from the full version. As its application is faster than the full version, it is useful to screen dynamic balance disorders¹⁴.

Both the BESTest and MiniBESTest were originally elaborated in English. Thus, they must be translated and cross-culturally adapted, and their psychometric properties must be assessed¹⁵. Therefore, the aims of the present study were to translate and conduct the cross-cultural adaptation of the BESTest and MiniBESTest to Brazilian Portuguese and to analyze their psychometric properties in older adults and individuals with PD.

Method

Participants

The study comprised a convenience sample including 35 older adults recruited in the community and 35 individuals with idiopathic PD diagnosed and recruited at the university outpatient clinic. The two groups included individuals from both genders. Nonetheless, volunteers with cognitive deficits on the Mini Mental State Examination (MMSE)^{16 17} or adverse clinical conditions, such as hemodynamic and/or clinical instability, were excluded. The group of older adults comprised individuals aged 65 years old or older and volunteers previously diagnosed with balance-affecting illnesses were excluded. The group of PD patients comprised individuals aged 50 years old or older, and those volunteers exhibiting other neurological diseases were excluded. The patients with PD were assessed within one hour from the intake of antiparkinsonian medication.

The study was approved by the Research Ethics Committee of the Universidade Federal de Minas Gerais - UFMG, Belo Horizonte, MG, Brazil and the participants signed informed consent forms.

Procedures and instruments of measures

Translation and cross-cultural adaptation

The process of translation and cross-cultural adaptation followed the method suggested by Beaton et al.¹⁸, which consists of six stages.

In stage one, the instruments were translated into Brazilian Portuguese by two independent bilingual translators who were native Portuguese speakers. In stage two, the two translations were synthesized into one consensus version, which was used as the basis for backward translation (stage three). This backward translation was performed by two independent translators who were native English speakers. These translators were blinded to the study aims and had no formal training in the health sciences. In stage four, the final version of the instruments was revised by a panel of experts, who were charged with consolidating the questionnaires and prepare the pre-final version¹⁸.

In stage five, the instruments were subjected to pretesting, during which they were applied to 10 older adults and 10 individuals with PD. The participants expressed no doubts as to the meaning of the BESTest items, and the responses showed no ambiguity; therefore, no changes had to be made to the translated version. In the case of the MiniBESTest, the examiner expressed doubts with the scoring of items seven and eight. The authors of the original version were contacted and recommended modifying the wording of the last sentence of the examiner's instructions in the corresponding items to "Include inclination or hip strategy as 'instability', and score one level below". Thus, the final Brazilian versions of the BESTest (^{Appendix 1}) and MiniBESTest (^{Appendix 2}) were concluded. Communication with the authors of the original instruments was maintained throughout the entire process to improve the final version (stage six)¹⁸.

First, clinical, demographic, and anthropometric data were collected. Next, the volunteers were evaluated with BESTest and MiniBESTest by a trained examiner. The disease severity of the individuals with PD was rated according to the Hoehn and Yahr Scale (HY)^{19 20}. Patients in stage 0 (zero) do not exhibit any sign of disease, those in stages 1 to 2.5 exhibit mild disability, those in stage 3 exhibit moderate disability, and those in stages 4 and 5 exhibit severe disability²¹. The individuals with PD were also assessed by the Motor Examination of the Unified Parkinson's Disease Rating Scale (UPDRS)²², which serves to assess the progression of PD and the effects of therapeutic interventions²².

Test-retest reliability

To test their reliability, the BESTest and the MiniBESTest were applied twice with a seven-day interval to the first 10 older adults and the first 10 individuals with PD included in the study. The reliability of the tests was assessed by the intraclass correlation coefficient (ICC).

Statistical analysis

Descriptive statistics were used to characterize the sample. The ICC was calculated using software SPSS for Windows version 17.0.

According to Portney and Watkins²³, the test-retest reliability is appropriate when ICC >0.75, whereas ICC<0.75 indicates poor to moderate reliability.

The final versions of the instruments were assessed using Rasch analysis, which is a probabilistic model widely used in the field of rehabilitation to evaluate the psychometric properties of scales²⁴. This analysis allows for the difficulty of the items and the abilities of the individuals to be calibrated on the same linear continuum. This linear continuum was divided into equal intervals (logits) along which each item and individual were aligned^{24 25}. The basic assumption of the present study was that the better the individual's balance, the greater his or her odds to score high in the investigated tests.

The analysis was performed using software Winsteps, version 3.72.3/2011²⁶, which calculates values such as MnSq and t in two formats: infit and outfit. The infit statistic is sensitive to the variation of the scores representing the subjects' abilities, and the outfit statistic reflects the occurrence of unexpected responses^{25 27}. These values are used to establish whether the items fit with the unidimensionality concept, with values of MnSq=1±0.4 associated with t=±2 being acceptable²⁷. Erratic scores represent a threat to the validity of a test and must be revised. The condition in which more than 5% of the items of a scale exhibit erratic scores indicate that the corresponding combination of items is not able to measure a unidimensional construct ²⁷, and a lower rate of erratic scores indicates that a particular set of items is able to define a dimension or construct^{24 25 27}. As the BESTest comprises 36 items (taking into consideration the tasks scored separately for the right and left sides of the body), up to two (1.8 rounded) items are allowed to not fit with the model.

The Rasch model further includes the person and item separation indices, which indicate the number of ability levels among which the individuals might be distributed and the number of difficulty levels among which the items are distributed, respectively^{24 25 27}. The number of levels among which the individuals and items could be distributed was calculated using the equation recommended by Bond and Fox²⁷. A satisfactory test divides the subjects into at least three ability levels: low, medium, and high²⁵. Ideally, the estimated values of reliability to calibrate the measures must be >0.80²⁷.

Results

Sample characterization

The average values of the parameters assessed in the group of older adults were the following: age=73.8±7.4 years old (65-97), body mass index (BMI) =25.6±3.7 (19.1-33.6), and MMSE=24.5±4.3 (13-30). Four volunteers were illiterate. In addition, most (18) of the volunteers had up to four years of schooling, eight had five to eight years, four had finished secondary school, and one had finished college. Of these 35 older adults, 26 were female, and one used a Canadian cane. In this subsample, 45.7% were married, 31.4% were widowers, 14.3% were single, and 8.5% were separated/divorced. Fourteen volunteers were sedentary, and 21 exercised using at least one of the following modalities: walking, dancing, yoga, and weight training.

The average values of the parameters assessed in the group of individuals with PD were the following: age=66.5±10.3 years old (50-88), BMI= 24.6±3.6 (17.6-32.9), MMSE=25.5±3.3 (16-31), UPDRS=20.2±8.8 (3-40), disease duration= 9.9±6.4 years (1-27), and HY stage 2.02±0.8 (1-4). Most participants (68.6%) were in HY stage 1-2, with 8.6% in stage 2.5, 20% in stage 3, and 2.9% in stage 4. Three volunteers were illiterate. In addition, most (12) of the volunteers had up to four years of schooling, seven had five to eight years, nine had finished secondary school, and four had finished college. Of these 35 participants, 14 were women, two used common canes, and one used a Canadian cane. In this subsample, 62.9% were married, 17.1% were widowers, 14.3% were separated/divorced, and 5.8% were single. Sixteen volunteers were sedentary, and nineteen exercised using at least one of the following modalities: walking, water aerobics, weight training, and T'ai Chi Ch'uan.

Test-retest reliability

The ICC values of both tests were adequate (>.90)²³. The ICC value of the BESTest in the population of older adults was 0.98 and was 0.92 in the individuals with PD. The ICC values of the MiniBESTest were 0.99 and 0.95, respectively.

Rasch analysis

In the BESTest, the calibration stability of the items and individuals was 0.97 and 0.95, respectively. The results of the Rasch analysis relative to the BESTest are described in Table 1, in which the values of the calibration or difficulty of items, MnSq and t (infit and outfit) are individually reported. The individual separation index was 4.19, which indicates that the items distributed the subjects among six balance ability levels. The item separation index was 5.36, which corresponds to approximately seven levels of difficulty.

Table 1
Calibration of BESTest items.

		Item	Calibration	Error	Infit		Outfit
					MnSq	T	MnSq	t
27.	SG	Timed "Get Up & Go" with dual task	2.63	.17	1.31	1.7	1.34	1.1
4.	BC	Hip/ trunk lateral strength	1.59	.16	1.19	1.2	1.13	.7
11.	TR	Stand on one leg - E	1.15	.16	.88	-.7	.92	-.4
11.	TR	Stand on one leg - D	1.10	.16	.89	-.7	.89	-.5
2.	BC	CoM alignment	1.07	.16	1.43	2.5	1.37	1.8
3.	BC	Ankle strength and range	1.07	.16	.89	-.6	.91	-.4
23.	SG	Walk with head turns *	1.05	.16	1.46	2.6	1.43	2.1
5.	BC	Sit on floor and standup	.79	.16	1.46	2.5	1.32	1.6
17.	RR.	Compensatory stepping correction - backward	.77	.16	.96	-.2	.92	-.4
21.	SG	Gait - level surface	.74	.16	.45	-4.2	.53	-3.1
24.	SG	Walk with pivot turns	.71	.16	1.29	1.7	1.28	1.5
19D.	SO	Eyes closed, foam surface	.66	.16	.73	-1.8	.79	-1.2
10.	TR	Rise to toes	.63	.16	.75	-1.6	.70	-1.8
7.	SL	Functional reach forward	.38	.17	.70	-1.9	.75	-1.4
1.	BC	Base of support *	.33	.17	1.62	3.1	1.89	3.8
8.	SL	Functional reach lateral - E	.32	.17	.70	-1.9	.85	-.8
15.	RR	In place response - backward	.30	.17	1.33	1.8	1.46	2.2
8.	SL	Functional reach lateral - D	.18	.17	.71	-1.8	.91	-.4
16.	RR	Compensatory stepping correction-forward	.18	.17	.89	-.6	.90	-.5
26.	SG	Timed "Get Up & Go"	.02	.18	.93	-.3	.80	-1.0
25.	SG	Step over obstacles	-.01	.18	.59	-2.6	.63	-2.1
6.	SL	Lateral lean - E	-.08	.18	.82	-1.0	.90	-.5
6.	SL	Lateral lean - D	-.11	.18	.89	-.5	.90	-.5
18.	RR	Compensatory stepping correction -lateral - D	-.11	.18	1.23	1.2	.95	-.2
18.	RR	Compensatory stepping correction -lateral - E	-.21	.19	1.46	2.2	1.23	1.1
12.	TR	Alternate stair touching	-.60	.20	.55	-2.6	.76	-1.1
19C.	SO	Eyes open, foam surface	-.77	.21	.74	-1.3	.77	-.9
14.	RR	In place response - forward	-.82	.22	1.16	.8	1.17	.7
22.	SG	Change in gait speed	-.97	.23	1.10	.5	.78	-.8
6.	SL	Sitting verticality - D	-1.12	.24	1.12	.6	1.25	.9
19B.	SO	Eyes closed, firm surface	-1.18	.24	.61	-1.9	.54	-1.8
6.	SL	Sitting verticality - E	-1.36	.25	1.10	.5	1.09	.4
13.	TR	Standing arm raise	-1.43	.26	.71	-1.2	.71	-.9
20.	SO	Incline - eyes closed	-1.57	.27	1.34	1.3	.82	-.4
19A.	SO	Eyes open, firm surface	-2.53	.36	.93	-.1	.42	-1.2
9.	TR	Sit to stand	-2.82	.40	1.27	.8	1.02	.3

• Erratic items: MnSq>1.4 and t>2. Sections of the test: BC: Biomechanical constraints; SL: Stability limits; TR: Transitions - Anticipatory postural adjustment; RR: Reactive postural responses; SO: Sensory orientation; SG: Stability in gait.

In Table 1, the items are listed in decreasing order of difficulty. Item 27, "Timed Get Up & Go With Dual Task", was the most difficult, and item 9, "Sit to Stand", was the easiest one. As a whole, the items were well distributed as virtually all the sections included items with low, medium, and high difficulty, except for the section "Biomechanical constraints", which contains items that are concentrated at the upper area of the continuum and that thus are primarily more difficult. Among the 36 items included in the test, two had erratic scores, i.e., did not fit with the model expectations, which were item 1, "Base of support", and item 23, "Walk with head turns".

Figure 1 represents a map that depicts the levels of difficulty of the items on the left and the continuum of the ability of the sample balance on the right, as measured by the BESTest. No items were aligned at the level of the individuals at either endpoint. Most items exhibited medium difficulty, whereas very easy and very difficult items, which might distinguish the individuals more accurately, were missing, although a ceiling effect did not occur. Most of the older adults exhibited medium to high abilities, and most individuals with PD exhibited medium abilities. Among the latter, those at the initial stages of the disease (HY=1-2.5; UPDRS=17.2±7.1) were located at the upper area of the continuum, which denotes better balance than the individuals in the moderate and severe stages (HY=3 and 4; UPDRS=30.4±5.7), who clustered towards the middle and lower areas, denoting worse balance.

Figure 1
Map representing the distribution of people and items from equilibrium measured by the BESTest. Each "X" on the left represents a test item, unidentified for easy viewing. The subjects are on the right: E=elderly and P=subjects with PD (shaded). The next number identifies the age, and the number after the dash indicates the HY classification of the individuals with PD. M is the average of the items on the left and right sides of the subjects.

In the Rasch analysis of the MiniBESTest, the individual separation index was 3.16, which indicates that the items distributed the subjects among approximately four levels of balance ability. The item separation index was 6.41, which corresponds to approximately nine levels of difficulty. The calibration stability of the items and individuals was 0.98 and 0.91, respectively. Similar to the full version, in the MiniBESTest, "Timed Get Up & Go With Dual Task" (item 14) was the most difficult, and "Sit to Stand" (item 1) was the easiest one. All the items fit with the model expectations (average calibration of 1.7 logits; average values of the subjects, infit [MnSq=0.98; t=0] and outfit [MnSq= 0.98 and t=0.2]); average values of the items, infit: [MnSq= 1.0 and t=0] and outfit [MnSq=0.98 and t=0]).

Figure 2 shows a map that depicts the continuum of the difficulty of the items on the left and the sample balance ability continuum on the right, as measured by the MiniBESTest. Similar to the full version, in this case, most items had medium difficulty, whereas items allowing the assessment of the individuals with very high or very low ability were lacking. Additionally, here, the individuals with PD exhibited a tendency to poorer balance than the older adults as the latter clustered in the upper area and the former in the middle area of the graphic. The individuals at the initial stages of PD (HY=1-2.5; UPDRS=17.2±7.1) tended to exhibit better balance and clustered in the upper area of the continuum, whereas those at the moderate and severe stages (HY=3 and 4; UPDRS=30.40±5.7) tended to exhibit poorer balance and clustered in the middle and lower areas of the graphic.

Figure 2
Map representing the distribution of subjects and items from equilibrium measured by the MiniBESTest. The numbers on the left represent the test items, and the subjects are on the right: E=elderly and P=individuals with PD (shaded). The next number identifies the age, and the number after the dash identifies the HY classification of the individuals with PD. M is the average of the items on the left and right sides of the subjects.

Discussion

The present study describes the versions translated and adapted to Brazilian Portuguese of new tools for balance assessment, i.e., the BESTest and MiniBESTest. The BESTest is the only test in which the items are categorized according to the balance control systems theory. This fact enables therapists to identify those individuals at risk for falls and to establish which factors contribute to the balance deficit to orient interventions^{8 28}.

The Rasch analysis performed in the present study showed a calibration stability of the items and individuals in both investigated tests, indicating that the measures were stable and reproducible. In the original versions, both instruments were considered to be reliable and valid to assess balance in individuals with PD^{9 14}.

Item 27, "Timed Get Up & Go With Dual Task", was considered to be the most difficult, which agrees with the results of another study²⁹. The performance of cognitive tasks (e.g., doing calculations and spelling words backwards) while walking is known to make the gait unstable³⁰, requiring a greater ability to perform the latter task. Item 9, "Sit to stand", was the least difficult and did not pose a challenge to balance in the present sample. The sample mostly comprised individuals with medium to high abilities and patients in the initial stages of PD, who were thus able to perform this task easily. Most items in the BESTest and MiniBESTest had medium difficulty (Figures 1 and 2, left side), and these tests lacked very easy and very difficult items, which would allow for the better assessment of the subjects, especially those with high performance levels. For this reason, the BESTest and MiniBESTest should be applied to populations other than the one included in the present study, such as youths and individuals with other diseases.

In the case of the BESTest, the Rasch analysis found only two erratic items (1 and 23); therefore, the combination of items was appropriate for the assessment of a unidimensional construct^{25 27}. Item 1, "Base of support" is scored by the examiner (presence or not of considerable deformities) and the subject (presence or not of pain in the lower limbs) together, which might account for the erratic scores. Pain is an individual and subjective experience that might be associated or not with actual or potential tissue damage, and it can become increased by negative emotional states³¹. However, chronic pain might also induce adaptive behaviors, leading the affected individual to not have signs^{32 34}. Therefore, examiners should pay special attention to individuals with depressive and negative behavior with regards to the pain. This observation should not be used to calculate the item differentially but to collect relevant information on the functional loss. In turn, item 23, "Walk with head turns", involved two simultaneous activities: walking in a straight line and turning the head. Therefore, it is a complex task that demands quick responses from the vestibular system¹, which might be affected by aging or disease^{1 3}. Individuals with PD have difficulty in performing simultaneous activities, and as a rule, they exhibit poorer performance in one of the tasks³⁵. Four volunteers in the present sample exhibited unexpected low scores in this item, including three older adults (O68, O69, and O78) and one individual with PD (P67; HY=2; UPDRS=25). The individuals were independent and had no dysfunctions; therefore, high scores were expected in item #23, similarly to what was seen in other items. These four volunteers, however, exhibited the minimum score because they stopped walking to turn the head to the side, which might represent a differentiated safety strategy. In addition, as Table 1 shows, item 23 is among the most difficult tasks. Therefore, the examiners should take the issues discussed here into account upon scoring this item.

The Rasch analysis showed a tendency in both the BESTest and MiniBESTest to distinguish between the groups of older adults and individuals with PD as a function of their distribution within the continuum. The older adults exhibited medium to high balance ability levels, whereas the individuals with PD exhibited medium levels. Other studies that compared the raw scores of the BESTest between the controls and individuals with neurological diseases⁸ and fibromyalgia²⁹ also found a poorer balance ability (lower scores) among the ill individuals^{8 29}.

The decisive factor in the distribution of the individuals with PD was the severity of disease. Most of the sample (77.1%) was at the initial stages of PD (HY=1-2.5) and clustered in the upper area of the continuum, whereas the participants at the moderate and severe stages, HY=3 and 4 (22.9%), clustered in the middle and lower areas of the continuum. Although the Berg Balance Scale is the instrument most widely used for the clinical assessment of balance, it exhibits a ceiling effect in milder cases of PD because those individuals attain the maximum scores^{12 13}. The results of the present study show that the BESTest and MiniBESTest exhibit a satisfactory ability to detect subtle balance deficits, thus making the distinction among individuals easier^{28 36}. This ability may be due to the fact that these tests assess gait and postural responses in addition to mobility (also assessed in the Berg Balance Scale), which might be affected at the earliest stages of disease¹³.

With regard to the clinical application of the BESTest and MiniBESTest based on the results of Rasch analysis, the main difference between both is that the former comprises more items and that, although most exhibit medium difficulty, these items cover a wider area of the ability continuum, resulting in a greater individual separation, as expected from an instrument used for diagnostic purposes. The item calibration is more widely spaced in the Mini version, resulting in less precise measures. These data correspond to the expected distinction between instruments designed for diagnostic and screening purposes, respectively. In the present study, the application of the BESTest lasted 25 minutes on average, varying from 21 to 31 minutes. In the study by Jones et al.²⁹, the average length was 17 minutes. The application of the MiniBESTest usually requires 12 to 15 minutes^{14 36}. The examiners should pay attention to the instructions described in the tests, especially for the items with a tendency towards variability in the responses.

The lack of frail older adults and the small number of individuals with advanced stages of PD might represent limitations of the present study. In addition, these tests must be applied to other types of populations in further studies for their validity to be more widely assessed.

The Rasch analysis showed that the set of items that compose the Brazilian versions of the BESTest and MiniBESTest exhibits appropriate construct validity, response stability, and discriminatory capacity among various balance ability levels in older adults and individuals with PD.

To the Conselho Nacional de Desenvolvimento Cientifico e Tecnológico – CNPq and Fundação de Amparo à Pesquisa do Estado de Minas Gerais - FAPEMIG, Brazil.

References

• ¹
  Horak FB, Macpherson JM. Postural orientation and equilibrium. In: Rowel LB, Shepherd JT, editors. Handbook of Phisiology. A critical, comprehensive presentation of physiological knowledge and concepts. New York: Oxford University Press; 1996. p. 255-92.
• ²
  Horak FB. Postural orientation and equilibrium: what do we need to know about neural control of balance to prevent falls? Age Ageing. 2006;35 Suppl 2:ii7-ii11. PMid:16926210. http://dx.doi.org/10.1093/ageing/afl077
  » http://dx.doi.org/10.1093/ageing/afl077
• ³
  Matsumura BA, Ambrose AF. Balance in the Elderly. Clin Geriatr Med. 2006;22:395-412. PMid:16627085. http://dx.doi.org/10.1016/j.cger.2005.12.007
  » http://dx.doi.org/10.1016/j.cger.2005.12.007
• ⁴
  Gonçalves DFF, Ricci NA, Coimbra AMW. Functional balance among community-dwelling older adults: a comparison of their history of falls. Rev Bras Fisioter. 2009;13(4): 316-23. http://dx.doi.org/10.1590/S1413-35552009005000044
  » http://dx.doi.org/10.1590/S1413-35552009005000044
• ⁵
  Borges LL, Garcia PA, Ribeiro SOV. Características clínico-demográficas, quedas e equilíbrio funcional de idosos institucionalizados e comunitários. Fisioter Mov. 2009;22(1):53-60.
• ⁶
  Tanner C, Hubble J, Chan P, editors. Epidemiology and genetics of Parkinson›s disease. In: Watts RL, Koller WC, editors. Neurologic principles and Practice. New York: McGraw-Hill; 1996. p.137-152.
• ⁷
  Lau LML, Breteler MMB. Epidemiology of Parkinson›s disease. Lancet Neurol. 2006;5:525-535. http://dx.doi.org/10.1016/S1474-4422(06)70471-9
  » http://dx.doi.org/10.1016/S1474-4422(06)70471-9
• ⁸
  Horak FB, Wrisley DM, Frank J. The Balance Evaluation Systems Test (BESTest) to Differentiate Balance Deficits. Phys Ther. 2009;89(5):484-498. PMid:19329772 PMCid:2676433. http://dx.doi.org/10.2522/ptj.20080071
  » http://dx.doi.org/10.2522/ptj.20080071
• ⁹
  Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142-148. PMid:1991946.
• ¹⁰
  Berg KO, Maki B, Williams JI, Holliday PJ, Wood-Dauphinée S. Clinical and laboratory measures of postural balance in an elderly population. Arch Phys Med Rehabil. 1992;70:1073-1080.
• ¹¹
  Tinetti ME, Inouye SK, Gill TM, Doucette JT. Shared risk factors for falls, incontinence and functional dependence: unifying the approach to geriatric syndromes. J Am Med Assoc. 1995;273:1348-1353. http://dx.doi.org/10.1001/jama.1995.03520410042024
  » http://dx.doi.org/10.1001/jama.1995.03520410042024
• ¹²
  Karuka AH, Silva JAM, Navega MT. Analysis of agreement of assessment tools of body balance in the elderly. Rev Bras Fisioter. 2011;15(6):460-6. PMid:22218711. http://dx.doi.org/10.1590/S1413-35552011000600006
  » http://dx.doi.org/10.1590/S1413-35552011000600006
• ¹³
  King LA, Priest KC, Salarian A, Pierce D, Horak FB. Comparing the MiniBESTest with the Berg Balance Scale to Evaluate Balance Disorders in Parkinson's Disease. Park Dis. 2012;2012. Epub 2011 October 24. http://dx.doi.org/10.1155/2012/375419
  » http://dx.doi.org/10.1155/2012/375419
• ¹⁴
  Franchignoni F, Horak FB, Godi M, Nardone A, Giordano A. Using psychometric techniques to improve the balance evaluation systems test: The Mini-BESTest. J Rehabil Med. 2010;42:316-324.
• ¹⁵
  Guillemin F, Bombardier C, Beaton DE. Cross-cultural adaptation of health-related quality of life measures: literature review and proposed guidelines. J Clin Epidemiol. 1993;46:1417-1432. http://dx.doi.org/10.1016/0895-4356(93)90142-N
  » http://dx.doi.org/10.1016/0895-4356(93)90142-N
• ¹⁶
  Bertolucci PH, Brucki SM, Campacci SR, Juliano Y. O mini exame do estado mental em uma população geral: impacto da escolaridade. Arq Neuropsiquiatr. 1994;52:1-7. PMid:8002795. http://dx.doi.org/10.1590/S0004-282X1994000100001
  » http://dx.doi.org/10.1590/S0004-282X1994000100001
• ¹⁷
  Brucki S, Nitrini R, Caramelli P, Bertolucci PHF, Okamoto IH. Sugestões para o uso do Miniexame do estado mental no Brasil. Arq Neuropsiquiatr. 2003;61(3-B):777-781. PMid:14595482. http://dx.doi.org/10.1590/S0004-282X2003000500014
  » http://dx.doi.org/10.1590/S0004-282X2003000500014
• ¹⁸
  Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the Process of Cross-Cultural Adaptation of Self-Report Measures. Spine (Phila Pa 1976). 2000;25(24):3186-3191. PMid:11124735. http://dx.doi.org/10.1097/00007632-200012150-00014
  » http://dx.doi.org/10.1097/00007632-200012150-00014
• ¹⁹
  Hoehn M, Yahr M. Parkinsonism: onset, progression and mortality. Neurology. 1967;17:427-442. http://dx.doi.org/10.1212/WNL.17.5.427
  » http://dx.doi.org/10.1212/WNL.17.5.427
• ²⁰
  Goetz CG, Poewe W, Rascol O, Sampaio C, Stebbins GT, Counsell C, et al. Movement Disorder Society Task Force Report on the Hoehn and Yahr Staging Scale: Status and Recommendations. Mov Disord. 2004;19(9):1020-1028. PMid:15372591. http://dx.doi.org/10.1002/mds.20213
  » http://dx.doi.org/10.1002/mds.20213
• ²¹
  Goulart F, Pereira LX. Uso de escalas para avaliação da doença de Parkinson em fisioterapia. Fisioter Pesqui. 2005;11(1):49-56.
• ²²
  Goetz C, Poewe W, Rascol O, Sampaio C, Stebbins GT. The Unified Parkinson's Disease Rating Scale (UPDRS): Status and Recommendations. Mov Disord. 2003;18(7):738-750. PMid:12815652. http://dx.doi.org/10.1002/mds.10473
  » http://dx.doi.org/10.1002/mds.10473
• ²³
  Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. 3rd ed. Upper Saddle River: Prentice-Hall; 2009. 892 p.
• ²⁴
  Kielhofner G. Research in Occuptional Therapy. Methods for Inquiry Practice. FA Davies; 2006. p.184-200.
• ²⁵
  Silverstein BF, Fisher WP, Kilgore KM, Harley JP, Harvey RF. Applying psychometric criteria to functional assessment in medical rehabilitation: Defining interval measures. Arch Phys Med Rehabil. 1992;73:507-518. PMid:1622298.
• ²⁶
  Linacre JM. Winsteps Rasch Measurement Version 3.72.3 [Software]. Available from: http://www.winsteps.com. 2011.
  » http://www.winsteps.com
• ²⁷
  Bond TG, Fox CM. Applying the Rasch model: fundamental measurement in the human sciences. Mahwal: Lawrence Erlbaum Associates; 2001. 288 p.
• ²⁸
  Leddy AL, Crowner BE, Earhart GM. Functional Gait Assessment and Balance Evaluation System Test: reliability, validity, sensitivity, and specificity for identifying individuals with Parkinson disease who fall. Phys Ther. 2011;91:102-113. PMid:21071506 PMCid:3017321. http://dx.doi.org/10.2522/ptj.20100113
  » http://dx.doi.org/10.2522/ptj.20100113
• ²⁹
  Jones KD, Horak FB, Winters KS, Morea JM, Bennett RM. Fibromyalgia is Associated with Impaired Balance and Falls. J Clin Rheumatol. 2009;15(1):16-21. PMid:19125137 PMCid:2836495. http://dx.doi.org/10.1097/RHU.0b013e318190f991
  » http://dx.doi.org/10.1097/RHU.0b013e318190f991
• ³⁰
  Hollman JH, Kovash FM, Kubik JJ, Linbo RA. Age-related differences in spatiotemporal markers of gait stability during dual task walking.Gait Postur. 2007;26(1):113-119. PMid:16959488. http://dx.doi.org/10.1016/j.gaitpost.2006.08.005
  » http://dx.doi.org/10.1016/j.gaitpost.2006.08.005
• ³¹
  Da Silva JA, Ribeiro-Filho NP. Avaliação e Mensuração de Dor - Pesquisa, Teoria e Prática. São Paulo: FUNPEC; 2006. 467 p.
• ³²
  Philips HC, Grant L. The evolution of chronic back pain problems: a longitudinal study. Behav Res Ther. 1991;29(5):435-441. http://dx.doi.org/10.1016/0005-7967(91)90127-O
  » http://dx.doi.org/10.1016/0005-7967(91)90127-O
• ³³
  Hall AJ, Kalra L. Measuring quality of life: who should measure quality of life. BMJ. 2001;322(7299):1417-20. http://dx.doi.org/10.1136/bmj.322.7299.1417
  » http://dx.doi.org/10.1136/bmj.322.7299.1417
• ³⁴
  Retinck ICM, Ketelaar M, Jongmans KJ, Gorter JW. Parents of children with cerebral palsy: a review of factors related to the process of adaptation. Child. 2006;33(2):161-169.
• ³⁵
  Berardelli A, Rothwell JC, Thompson PD, Hallett M. Pathophysiology of bradykinesia in Parkinson's disease. Brain. 2001; 124: 2131-46. PMid:11673316. http://dx.doi.org/10.1093/brain/124.11.2131
  » http://dx.doi.org/10.1093/brain/124.11.2131
• ³⁶
  Leddy AL, Crowner BE, Earhart GM. Utility of MiniBESTest, BESTest and BEStest sections for balance assessment in individuals with Parkinson's Disease. JNPT. 2011;35(2): 90-97. PMid:21934364 PMCid:3178037.

Appendix 1 Translated version of BESTest to Brazilian Portuguese.

Appendix 2 Versão traduzida para o português-Brasil do MiniBESTest.

Publication Dates

• Publication in this collection
  June 2013

History

• Received
  30 Mar 2012
• Reviewed
  23 Aug 2012
• Accepted
  20 Jan 2013