ARTIGO ORIGINAL ORIGINAL ARTICLE

It is estimated that in developed countries between 1% and 2% of total healthcare expenditure goes on asthma treatment. The natural progression and causes of severe asthma are still not well defined. Severe persistent asthma is a less common situation, accounting for 5% of total asthma patients. Despite representing a small percentage of total patients, this group is largely responsible for the high health costs of asthma. Consequently, the improvement of risk factors can lead to both socioeconomic and health gains. The aim of this study was to examine risk factors for exacerbations in a group of patients with severe persistent asthma. It was a retrospective study, based on patient case histories that included subjects with severe persistent asthma with follow-up in an allergy unit be


INTRODUCTION
Horak et al. 1 considered balance as a nervous system skill to detect any instability either in advance and immediately and to generate coordinated responses that could restore the supporting base of body mass core, preventing falls. Effective maintenance of balance involves a number of central nervous system (CNS) and peripheral nervous system (PNS) structures 2 . According to Woollacott 3 , vestibular system is one of the main structures to maintain balance, given that it is considered as an absolute reference in relation to the others that also participate in this function, such as visual and somatosensorial systems.
When the set of visual, labyrinthic and proprioceptive information is not properly integrated by the CNS, there is disturbance in balance status, which can be manifested by body imbalance, and may lead to falls 4 . Aging may be responsible for these disorders. The elderly may have difficulty to precisely regulate these stimuli, which can be improved through specific training programs. Considering that daily activities are present in innumerous conditions that involve sensorial redundancy, skills to analyze and select information are essential to prevent falls 5 .
Systems in our body have physiological reserves that are characterized in the nervous system by their capability to reorganization, known as neuroplasticity 6,7 . As a result of aging, reserves are reduced, but not depleted, therefore, the creation of an ideal environment for motor learning may determine a significant improvement of the function 8 . Pohl and Winstein 9 stated that practice improves neural processing skills in the elderly as well.
Vestibular exercises, such as the ones described by Cawthorne and Cooksey, may serve as support for new arrangements of peripheral sensorial information, allowing new vestibular stimulation patterns necessary for new experiences to become automatic. This practice of balance would be capable of promoting improvement in reactions of balance and, consequently, reduce falls 10 .These exercises are part of a vestibular rehabilitation program and involve head, neck and eye movements, posture control exercises in different positions (seated, in two-leg and one-leg positions, walking), use of soft surface to reduce proprioceptive input, and exercises with closed eyes to exclude visual cues.
The purpose of the present study was to check a specific therapeutic approach for the vestibular system by the application of Cawthorne and Cooksey exercises, to see if they generated motor learning and contributed to improving balance and reducing the possibility of falls in the elderly.

MATERIAL AND METHOD
The research project was approved by the Research Ethics Committee, Universidade Castelo Branco, according to Resolution 196, 10/10/1996, National Health Council.

Population and sample
Elderly female patients, members of a district association in Barra Mansa/RJ, aged 60 to 69 years, were randomly chosen. All women aged 60 to 69 enrolled in the association were invited to be part of the study. The control group and the studied group were formed in a first come basis, and the 15 first applicants formed the studied group, whereas the other 15 formed the control group. They signed the informed consent term and agreed to join the study. The study was conducted between March and June 2003.
Female subjects were chosen because falls are more prevalent in women 4,11 . The age range represented the first decade of aging, when responses to therapeutic interventions and physical activities are more marked.
Exclusion criteria were presence of neurological, ENT, vascular, metabolic, degenerative or neoplastic disorders, which are confirmedly known to cause balance disorders.

Measurement instrument
Balance was assessed using Berg Balance Scale (BBS) (Annex A). This instrument is used to assess balance and risk of falls in elderly and takes into account the effect of environment. According to Gill et al. 12 , most of the falls in the elderly occur in everyday situations, especially in unfavorable environmental conditions. This scale uses 14 tests to assess subjects' skill to sit down, stand up, reach, turn around them, look over the shoulders, stand on one foot, and go upstairs. Total score is 56 and any rate equal or below 36 is associated with 100% risk of falls 13,14 .
This instrument shows excellent reliability (0.96) and moderate to high correlation with other balance functional assessment instruments, such as Barthel Mobility Scale, 0.67; Up and Go Test, 0.76; Tinetti Balance Scale, 0.91 15 . The scale has excellent test-retest objectivity (ICC = 0.98) 16 .
Absolute scores obtained in BBS were applied to reach the rate of Likelihood of Fall (PQ) using the following equation: 100% x exp (10.46 -0.25 x BBS score + 2.32 x history of instability) / [ 1 + exp (10.46 -0.25 x BBS score + 2.32 x history of instability)], and BBS score is the score obtained by the subject in the BBS. History of instability receives value 0 if there is no report of instability and value 1, if there is report of instability.

Procedures
Out of forty-five subjects that responded the invitation, 33.3% (15 subjects) did not participate in the study because they did not comply with the inclusion criteria or were not interested. The percentages were: 46.6% (7 subjects) presented metabolic disorders (diabetes mellitus); 6.6% (1 subject), vascular disorder (uncontrolled high blood pressure); 13.3% (2 subjects) had neurological affections (sequelae of  13. Stand without support one foot in front of the other Instructions: (Show the subject). Place one foot in front of the other. If you do not manage to have one in front of the other, take the longest step you can to have your toes touch your ankle. Score: Check the category that applies.
(1) (0) manages to place the manages to place the feet manages to take a small needs help to take loses balance feet correctly and correctly and step independently and the step, but is able when takes a independently and independently and maintains the position to maintain the step forward maintains the position maintains the position for for 30 seconds position 15 or stands up for 30 seconds 30 seconds seconds.
14. Stand up on one foot Instructions: Stand up on one foot as long as you can, but without support. Score: Check the category that applies.
(1) (0) manages to raise the leg manages to raise the leg manages to raise the leg tries to raise the leg; unable does not independently and independently and independently and maintains to maintain the position for manage to try or maintains the position maintains the position the position for over 3 3 seconds, but remains needs help not for over 10  cerebral vascular accident); 40% (6 subjects) had ENT affections (labyrinthitis); 13.3% (2 subjects) had significant visual deficit, and 13.3% (2 subjects) were able to participate but did not show interest. The sum up of percentages is higher than 100% because subjects could have manifested more than one disorder. Thus, we selected 30 subjects that participated in the study. They were all assessed by BBS and after nine weeks, they were reassessed. Components of the studied group were submitted to vestibular stimulation with Cawthorne and Cooksey exercises, three times a week, for sixty minutes during this time interval (Annex B). The interval represented the mean time recommended to assess the progression of a patient submitted to vestibular rehabilitation 17 .
All participants presented a percentage of attendance to the program sessions that was equal or higher than 75%, so that data could be included in the analysis.

Data analysis
Initial and final data were analyzed and compared by two statistical tests of significance: T-Student test and Wilcoxon test. T-Student test requires variables to be distributed as regular likelihood. To check this assumption we used Kolmogorov-Smirnov test. Level of significance (a) was 5%, that is, p<0.05 18 .

RESULTS AND DISCUSSION
Mean age of the studied group was 64.8 years (±2.95) and to the control group it was 65.46 years (±2.85).
Subjects presented a first assessment of BBS scores with the respective PQ showed in Table 1. The percentage of subjects in each PQ value is shown in Graph 1. No elderly subject presented PQ equal to 100%, which would be expectable in a survey with healthy elderly, but 24 of them (79.8%) presented PQ between 28% and 73%, which demonstrates that subjects without diseases present likelihood of suffering falls that is high enough to restrict their daily life activities. The process of aging, per se, determines gradual system failure, regardless of the presence of disorders, but there was a minimum of general BBS score, which inevitably occurs during the process, that is significant considering non-linear progression. Shumway-Cook et al. 15 reported that 25% to 35% of the population aged over 65 years tends to suffer falls.
Data presented in Graphs 2 and 3 detected homogeneity between both groups in the 1st assessment, because there were no statistically significant differences in results (p>0.05).
This study managed to demonstrate that BBS is sensitive to detecting abnormalities in balance of healthy  elderly. The elderly in the studied and control groups did not present statistically significant difference when assessed for the first time (p>0.05), but for the second assessment we observed statistically significant differences between the two groups (p<0.005), provided by significant improvement in balance (p<0.05) in the studied group after intervention (Table 2 and Graph 2). There was significant improvement in PQ (p<0.05) for the studied group, with reduction of 30.4% in the likelihood of having falls (Table 3 and Graph 3), leading to the conclusion that a specific intervention, based on stimulation of specific system, generated positive functional responses in this studied group. The clinical importance of this result lies in the fact that falls are one of the main factors that contribute to morbidity and mortality of the elderly. Thus, preventing falls by improving balance provides basic conditions for the maintenance of physical independence. Differences between the control and studied group to PQ both in the first and second assessment were not significant (p>0.5 and p>0.05, respectively), despite the tendency towards increase in difference between both groups in the 2nd assessment, which means that significant improvement obtained in the studied group, confirmed in this study, may not be extrapolated to other groups (Graph 3).
Data in Tables 4 and 5 and in Graphs 2 and 3 show, respectively, that the control group did not present significant differences between means of BBS (p>0.05) in the 1st and 2nd assessments nor between PQ means (p>0.05).

CONCLUSIONS
The results found in this study confirmed that according to BBS, healthy elderly subjects have balance disorders and run the risk of falling.
Cawthorne and Cooksey exercises applied as described in the procedures were capable of improving balance in our sample, consequently reducing the likelihood of fall.
The results and conclusions described here confirmed the expectations of different authors [19][20][21][22] that suggested improvement in balance and in likelihood of falls when there was application of vestibular stimulation in healthy elderly, given that these subjects have really presented significant improvement.   Elderly subjects, who reported or not presence of posture instability and/or the event of fall, should be submitted to vestibular stimulation exercises, exercises that are easy to apply and affordable, which are preventive and curative concerning balance deficits and risk of falls. Considering that falls are aspects that substantially change the quality of life of the elderly and that life expectancy of the population in general has increased significantly, leading to increasingly higher elderly population every year, general therapeutic interventions directed to the elderly and specially those that provide prevention of falls owing to improvement of posture stability, will eventually lead to improvement in quality of life of this part of the population, which is currently the priority of any and all health policies. Table 2. Berg Balance Scale by subject in the studied group in the 1st and 2nd assessments (pre and post-intervention). Subjects  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  1st assess.  55  52  52  56  51  55  54  54  56  54  55  49  52  49  54  2nd assess.  56  53  56  56  56  56  55  55  56  55  56  56  53 53 55 X1 = 53.2 dp = 2.27 T-Student t = -3.6495 Wilcoxon Z = -2.6858 X2 = 55.1 dp = 1.18 (values) p = 0.0026 (values) p = 0.0072 Table 3. Values of likelihood of fall by subjects in the studied group in the 1st and 2nd assessment (pre and post-intervention).