Chapter Five: Discussion
The study findings will be discussed in this chapter. Tables and diagrams from the results section will be described. In addition, the relationship between cause and effect will be determined whenever it is possible. Finally, the limitations of the study, its recommendations and the conclusion will be given at the end of the chapter.
Three studies, which were not presented in the literature review chapter, will be introduced in this chapter to describe BMI and its relationship with the ability to balance (Kuczmarski and Flegal, 2000; Deforche,2008; Ku, 2012)
5.1 Description of the study findings
a. Demographic data
Table 1.1 shows the mean and standard deviation of age, weight, height and BMI for the 21 recruited participants. It was calculated that age had a mean of 23.43 and a standard deviation of 4.094, so the findings are only pertinent to this particular age range. In addition, a mean of 76.92 and standard deviation of 15.22 was calculated for weight. It was notable that the standard deviation for weight is quite high, which will affect the results obtained. Despite the standard deviation of the weight being high, the BMI standard deviation of 4.49 seems to be appropriate since it will not create a variation in the data obtained. Leg dominancy was also determined, as shown in Table 1.2. It was found that four participants had a dominant left leg; the other 17 participants had a dominant right leg.
b. Main findings
Table 2.1 and Figure 3.1 showed the mean of AP-EXC, ML-EXC, AREA and TOTAL-EXC before any intervention was performed (baseline measurement). In fact, there was a slight increase in the mean of the sitting still group compared to the task-oriented training group. Accordingly, the differences between the mean of the sitting still group and the task-oriented training group were 0.087, 0.035 and 0.11 points for AP-EXC, ML-EXC and TOTAL-EXC, respectively. In addition, the difference in AREA mean between both groups was 3.92 point, which is considered to be high and may have affected the data obtained after intervention.
Regarding the post-intervention measurements for both groups, the results were explained and detailed in Table 2.2 and Figure 3.2. It was found that there was a reduction in the mean of the task-oriented training group as well as the sitting group when compared to the baseline measurement. This finding indicates that there was an improvement for both groups after intervention. This reduction reflects a reduction in COP movement in almost all directions. The improvement in the task-oriented training group may be attributed to exercise performance; however, the learning effect may have contributed to the improvement in the sitting still group.
Moreover, the change score measurements for both groups were described in Table 2.3 and Figure 3.3. The results showed that the mean of AP-EXC, ML-EXC, AREA and TOTAL-EXC for the sitting still group was lower than the mean of the four measurements in the task-oriented training group. These results may be due to the mean score baseline measurements of the sitting group being higher than those of the task-oriented training group, making the change observed in the mean score larger.
Figure 3.4 showed the differences between baseline measurements and post-intervention measurements for the task-oriented training group. There was a reduction in the mean score that ranged between 0.026 and 0.076. Nevertheless, the mean of the AREA measure after engaging in the interventions has been increased by 1.33 point. This increase in the AREA mean score may have resulted from an extreme score influence. On the other hand, the data obtained from Figure 3.5 concerning the sitting still group suggests a reduction in the mean score when comparing the baseline with the post-intervention measurements, where the reduction ranged between 0.11 and 3.21 points. Accordingly, the sitting still group demonstrated more improvement in mean score when compared to the task-oriented group. As mentioned, the higher change can be attributed to the higher baseline measurement.
An important finding should be discussed: namely, the variance of scores that reflected how data is dispersed and how it appears in the current data as standard deviation. If the standard deviation (variance) is small, the data will be more similar. It is essential to highlight that an increase in standard deviation will result in a skew of the mean, thus affecting the accuracy of the data (Hicks, 2009). The low standard deviation presented in Tables 2.1 and 2.2 and 2.3 of AP-EXC, ML-EXC and TOTAL-EXC for baseline measurements, post-intervention measurements and change score measurements for both groups indicates that the data were close to the mean, leading to a more reliable result. On the other hand, the high standard deviation of AREA for baseline measurements, post-intervention measurements and change score measurements for both groups suggests that the data spread widely from the mean, thus affecting the reliability of the results. The standard deviations of height, age and BMI, as presented in Table 1.1, were low, while the standard deviation of weight was high, indicating a great variance liable to affect the accuracy of the results. Although the standard deviation of the BMI was low, the raw data presented in Appendix F demonstrates that eight participants had a BMI ranging from 25 to 29 kg/m2 and four participants had a BMI ≥ 30. According to Kuczmarski and Flegal (2000), individuals with a BMI in the range from 25 to 29 kg/m2 are considered to be overweight and those with a BMI ≥ 30 are considered to be obese. Deforche (2008) has found that being overweight will result in a reduction of static and dynamic balance capacity as well as postural skills. Also, Ku (2012) emphasised the importance of the proper selection of individuals with appropriate weight, especially when assessing static balance. Indeed, the BMI spread may have affected the data obtained, leading to a reduction in data quality.
In spite of the slight improvements in the mean and standard deviation, the data represented in Tables 3.1, 3.2 and 3.3 shows the results for a paired T test for baseline measurements, post-intervention measurements and change score measurements, suggesting that the task oriented training had no effect on balance performance in healthy individuals. Since the significance was set to be ≤ 0.05 for the study and the significance scores of all measurements were ranged between 0.084- 0.549, therefore there were no significance differences between the task-oriented group and the sitting still group.
5.2 Previous research and relevant finding
The results of the study have similarities with Outermans (2010) who assessed the effect of four weeks of high intensity task-oriented training on the gait ability of 44 stroke patients. The experimental group received high intensity task-oriented training that consisted of 10 exercise stations as well as walking training. The control group received low intensity physiotherapy. Outermans’s (2010) study indicated that high intensity task-oriented training improved walking capacity and walking speed. However, there was no improvement in balance control. Outermans (2010) attributed this result to the lack of Berg balance scale sensitivity when the baseline measurements are high. Also, the study indicated that the influence of high-dose practice on balance ability would be lower than effort-related measures such as walking speed and walking distance.
In contrast to the current study finding which stated that there was no significant difference between both groups; Salbach (2005) conducted a randomised controlled trial to examine the effectiveness of task-oriented training on balance self-efficiency in 91 post-stroke patients. The intervention-group participants were involved in a progressive programme, which included performing 10 tasks. While the control-group subjects received upper limb tasks while sitting. The finding suggested that six weeks of task-oriented training produced a significant improvement in balance self-efficiency. Another study with similar findings was conducted by Katz–Leurer (2009). It evaluated the efficacy of a home programme of task-oriented training on improving balance ability in cerebral palsy and traumatic brain-injury patients. The study concluded that the training increased functional balance performance and the improvement was maintained even after the end of training. However, muscle strength and walking ability didn’t improve.
Kim (2012), who investigated the efficacy of task-oriented training in enhancing trunk-control ability, balance performance and gait ability in stroke patients, obtained some promising results. The study outcomes measures, which consist of trunk impairment scale, time up and go test and berg balance scale and the finding of these outcome measures revealed a significant improvement for the task-oriented training group when compared with the control group. As a result of this, Kim (2012) showed a slight difference from Salbach (2005) in terms of the effect of task-oriented training on gait ability.
A study by Yang (2006) that assessed the effect of task-oriented training on muscle strength and functional performance in stroke patients was in agreement with a previous study done by Kim (2012). Participants in the experimental group engaged in task-oriented training for four weeks, while no rehabilitation training was provided for the control group. The results showed an improvement in muscle strength for the task-oriented training group of between 23.9% and 36.5% for the non-paretic side, and 10.1% and 77.9% for the paretic side. The control group, by contrast, experienced a decline in muscle strength. Also, the data for the experimental group revealed a significant improvement in all functional tests, except the step test. While the results for the control group indicated that there was either no change or a decline in all functional tests. Accordingly, the study concluded that task-oriented training improves the muscle strength of the lower limbs and the improvement may be transferred to functional abilities.
Salem (2009) examined the ability of task-oriented training to enhance the functional ability of cerebral palsy patients. It was found that five weeks of task-oriented training enhanced dimensions D and E (p=0.009) of gross motor function measure (GMFM). Furthermore, it took the experimental group less time to complete the timed up and go test (p=0.017). Consequently, task-oriented training is considered to be beneficial for improving functional ability.
The disparity in the results of the studies may be due to several factors, such as: study sample, outcome measure sensitivity, intervention goal and intervention components. In fact, current study results may be affected by numerous factors and these may affect the accuracy of the data obtained.
One of the factors that emerged that may have contributed to the data obtained is the use of healthy participants. As already mentioned, using healthy participants may have a role in determining baseline data for other studies as well as obtaining data that will not be distorted by the symptoms of a different pathology . However, studies that used healthy participants followed correlation and test-retest designs. In Pinsault (2008), Bauer (2008) and Geurts (1993), healthy participants were recruited to determine the reliability of force-platform measurements in assessing postural control. The results of these studies are promising and may have an influence on both research and rehabilitation. These studies have succeeded with the selection of healthy participants and provided good results because it wasn’t necessary to present a cause and effect relationship. Also, the effect of specific training on participant performance was not needed in this type of study. In fact, studies that investigate specific training and required a cause and effect relationship to be demonstrated require participants with various pathologies.
Salbach (2004) and Outermans (2010) recruited post-stroke patients in order to evaluate the effect of task-oriented training. Salbach (2004) recruited participants within one year of a stroke, whereas Outermans (2010) recruited participants just two to eight weeks after stroke onset. Results that support the use of task-oriented training were found in both studies. Obviously, different deficits and impairments were presented among participants in Salbach (2004) and Outermans (2010). Nevertheless, these impairments helped to clarify the effectiveness of training to improve their conditions, and therefore clarified the success of the training programme. Other studies recruited participants with different levels of GMFM (Salem, 2009; Katz-leurer, 2008). Accordingly, the disabilities associated with each level create a great spectrum that helps to determine the influence of task-oriented training.
In addition, the interpretation of study results requires a consideration of the sample sizes. It is noted that the power of a study depends on raw sample size: the larger the sample size, the greater the study’s power (Portney and Watkins, 2009). In fact, this study has a small sample size, involving just 21 participants, with the absence of a power calculation. As a result, the study may be insufficient to determine the difference between the groups, and the elimination of type I and II errors was much harder. It is true that the sample size of the current study was dependent upon the power calculations from previous studies; however, performing a power calculation may increase the reliability and accuracy of the chosen sample size.
Several studies used a relatively large sample size, ranging from 44 to 91 participants (Salbach, 2005; Yang, 2006; and Outermans, 2010). Such sample sizes increase the chance of showing the effect of training by increasing the ability of determining the difference between the groups, and thus being able to obtain a significant result. It is worth mentioning that some studies have a small sample size of between 10 and 20 participants (Salem, 2009; Katz-leurer, 2009; and Kim, 2012). Nevertheless, the result shows significant improvements when comparing the experimental group with the control group. Although the studies had small sample sizes, they were able to produce the desired results. It is therefore important to realise that sample size alone can’t be relied upon; it is necessary to highlight other elements contributing to the results, such as the instrumentation used and the intervention components.
With regard to the instruments used, a force platform may not be sensitive enough to detect changes in small groups of participants with no disorders affecting their balance ability. Various studies that assessed the ability of force platforms to determine postural status involved different participant characteristics that may have affected the results obtained. Pajala (2008) suggests that force platforms can predict the risk of a fall in older adults without apparent disorders. Participants in this study had ages ranging from 63 to 76. Even though there was no apparent disorder in balance performance, the changes in balance ability related to the ageing process may, debatably, affect the accuracy of the result. In a study completed by Bauer (2008), which recruited participants of over 62 years, force platform measurements with ICC ≥ 0.90 produced results that were highly reliable. As mentioned, this result may be attributed to the deterioration of balance performance due to age. On the other hand, a study undertaken by Karlsson (2000) gave a result that supports the use of force platform measurements. In this study post stroke patients were recruited. The damage that occurs as a result of a stroke affects some components of balance. As a result, the accuracy of the result is reduced since the impairment associated with stroke onset increases the sensitivity of a force platform to detect change. A systematic review done by Piirtola (2006) revealed that force platforms have some predictive value for falls, but there is still insufficient research so it is difficult to reach an accurate and comprehensive conclusion.
In order to avoid the problems associated with the sensitivity of force platform measurement when detecting change in asymptomatic healthy participants, force platforms should be used in combination with other outcome measures. Most of the studies used multiple outcome measures to ensure that a slight change in participant performance would be detected (Bunchner, 1997; Schlicht, 2001; Hirsch, 2003; Salbach, 2004; Yang, 2006; and Outermans, 2010). Yang (2006) used the six-minute walk test, timed up and go test and step test to assess the functional ability of participants. His findings indicated that significant improvement was found in the six-minute walk the test and timed up and go test. However, the step test didn’t reveal any improvement. The study attributed the result of the step test to other components that affect stepping ability in combination with muscle strength. Using several outcome measures allowed the study to overcome the limitation of the step test, and detect the change between the groups and reach the desired result. Outermans (2010) used the six-minute walk test, ten-minute walk test, berg balance scale, and functional reach test as the study outcome measures. The six-minute walk test and the ten-minute walk test showed significant differences between the groups. No difference was found in either the berg balance scale or the functional reach test. The limitation of the berg balance scale affected the result obtained. Accordingly, the accuracy and reliability of data obtained may be enhanced when using different outcome measures.
Another aspect of the force platform must be taken into consideration: the standardisation procedure. As mentioned, determining the number of trials for each task is an essential step since it has a large effect on the results obtained (Duarte, 2010). The number of trials recommended for each task ranged from two to four. The current study performed one trial for each task since no more is considered feasible for MSc research. Also, the choice of the number of trials was based on a previous study done by Pajala (2008) and the result of the study indicated that the force platform test is a valid tool in determining postural control status. However, other studies suggest that three trials for each task may ensure more reliable data for force platforms (Pinsault, 2008, Bauer, 2008). In fact, performing three trials for each task may increase the opportunity for force platform measurements to detect changes in asymptomatic healthy participants.
Additionally, choosing the appropriate test for measuring balance ability is a major step in achieving the preferred results. Balance is considered to be a complicated skill that requires processing information from sensory, motor and biomechanical components (Guskiewicz, 1996). It is believed that several divisions fall under the concept of balance: postural control on a stationary basis, on voluntary movement, on involuntary movement and on external disturbance (Hummer, 2008). Thus, the balance outcome measures used in a study must be able to identify both static and dynamic elements in order to reflect the full status of balance ability. A force platform can be used to determine both static and dynamic balance (Guskiewicz, 1996). However, in the current study, participants were asked to perform single-leg standing on the force platform. In this way, the test only identifies the static component of balance. Therefore, the changes in balance ability will not be fully covered and determined.
Furthermore, the major challenge in conducting a study that ensures an accurate result is building an appropriate intervention programme. Task-oriented training is essential in a motor learning process. Also, a meaningful task allows greater functional organisation to take place (Salem, 2009). Thus, in order to reach the optimal learning effect, the task should be similar to real-life activities. The result of previous studies indicates that task oriented training provides a significant and long-lasting improvement to the cortical organisation (Salem, 2009). Moreover, recent evidence shows that task oriented training enhances motor function.
As mentioned, the optimal effect will take place when the tasks are specific and the training is strong and demanding enough to promote change. The intervention of current study involved practising the repetitive functional tasks found in everyday activities. Since the intervention characteristic and component contributed to the result, the choice of appropriate intervention parameters must be considered.
Training parameters should reflect appropriate difficulty and demands in order to create change in performance. The difficulty of training may be determined by the number of times an exercise is repeated, the number of exercise stations, as well as the number of training sessions. Salbach (2004), Salbach (2005), Yang (2006) and Outermans (2010) provided six to ten exercise stations for four to six weeks. The length of time each exercise was repeated at each station ranged from 2.5 minutes to 5 minutes, but the exact number of repetition for each exercise was unclear. Also, Outermans (2010) determined the progression of the exercise by increasing the repetition number and workload, depending on the heart rate. The results of these studies were positive and recommended the use of task oriented training. In contrast, the result of the current study suggested no difference between the task oriented training and sitting still groups in relation to balance performance. The repetition number in the current study of ten was chosen according to Nicholas’s (2009) recommendation. The number of exercise stations was three: reaching outside BOS, walking on balance beams and crossing over obstacle courses. These stations were determined on the basis previous studies (Kim, 2012; Outermans, 2010). Only one session of task oriented training was provided for participants. The choice of the number of exercise stations, as well as the number of training sessions was restricted by what is feasible for MSc research. It is important to mention that the repetition number, number of exercise stations and number of training sessions in the current study may not have been enough to promote change in balance ability among healthy participants.
One of the other elements that interfered with the results was rest period. Kim (2012), Kaesler (2007) and Hirsch (2003) provided a rest period for participants between each set of exercises ranging from 30 seconds to 2 minutes. The importance of providing a rest period in the current study emerged from its ability to reduce fatigue effect, which may increase body sway thereby affecting the data. The current study used a rest period of one minute between a set number of exercise repetitions (3 repetitions) and two minutes at the end of all the exercises. The length of the rest period was chosen on the recommendation of Nicholas (2009) who stated that motor performance exercise that used multiple joints should follow the rest period of exercise with a heavy or moderate load lasting from between one and three minutes. Although the study provided rest periods, they may not have been sufficient to reduce the fatigue effect. Therefore, the modification of the rest period may have enhanced the data obtained
5.3 Study limitations
When reviewing the methodological procedure of the study and the results it produced, several limitations have been identified. To start with, blinding the participants was not possible due to a study design that required all participants to be engaged in both groups (task oriented training and sitting still). Therefore, the data obtained may have been influenced by participants’ thoughts and reactions.
Besides not blinding the participants, recruiting healthy subjects created a great challenge to determining any change in balance performance since the group of participants didn’t have any pathology that caused a reduction in balance ability. In addition, the force platform’s sensitivity with asymptomatic healthy participants is questionable since its ability to reveal change in balance ability after each intervention was limited. Thus, the accuracy of the obtained data is reduced and the study’s results can’t be transferred to the general population.
Furthermore, the sample size was too small to reveal any differences between the two groups. In fact, the recruitment of 21 participants puts the results under the risk of type I and II errors since a larger sample size would have ensured the elimination of error. As a result, the study’s findings will not be generalised to the wider population.
Finally, various aspects of the trial are open to criticism, such as the rest period, the number of exercise repetitions and the number of training sessions. It worth mentioning that these elements may not have been strong and demanding enough to create a detectable change in balance ability.
5.4 Recommendation for future research
As previously mentioned, the current study contains many limitations which call the results obtained into question. Therefore, future studies should strive to overcome these limitations to achieve a better result.
To overcome the risk of type I and II errors and increase the generalisability of the study, larger sample sizes are needed. Additionally, in order to allow proper detection of the change between groups as well as the increase in sensitivity of the force platform, participants with pathologies that weaken their abilities need to be recruited, such as patients suffering from strokes, Parkinson’s disease and other neurological deficits that affect balance performance.
Furthermore, future studies should use the force platform in combination with other outcome measures in order to decrease the bias that resulted from not blinding the subjects. It will also increase the chance of detecting change in balance performance as the instrument used in the current study only focused on the component of static balance.
Moreover, increasing the rest period between each set of exercises from one to two minutes, and at the end of the training from two to three minutes may reduce the fatigue effect. In fact, increasing the number of repetitions as well as the number of training sessions to provide stronger intervention will likely produce notable differences between the groups.
5.5 Conclusion
In conclusion, interpreting the study results revealed that there was no significant difference between the task oriented training and the sitting still groups in the baseline measurement, the post intervention measurement or the change score measurement. This result can be attributed to several factors, such as the sample size, the characteristic of participants, the sensitivity of the force platform and the intervention elements. The study acknowledged these limitations and insisted on the need of future research to ensure a more accurate and definitive result.
Are you looking for a similar paper or any other quality academic essay? Then look no further. Our research paper writing service is what you require. Our team of experienced writers is on standby to deliver to you an original paper as per your specified instructions with zero plagiarism guaranteed. This is the perfect way you can prepare your own unique academic paper and score the grades you deserve.
Use the order calculator below and get started! Contact our live support team for any assistance or inquiry.
[order_calculator]