When the cognitive function of the 3 groups was examined, the LEF group displayed a similar MoCA score average as the HC group, despite the lower executive function. Both groups were found to have a significantly higher score than the IC group, showing that the cognitive test score of the LEF group is not a good indicator of executive function deterioration. The demographic variables showed no difference between the three groups, indicating that an adjustment for covariates is not needed when comparing these groups’ functions.
When analyzing the gait variables, LEF and IC had mostly similar performance, with the exclusion of the single limb support portion where there was a significant difference between these two groups. The gait ability of HC was found to be significantly greater than that of IC in the case of velocity, cadence, stride duration, stance duration, loading response phase, double limb support phase, and swing phase. Among these variables cadence, stride duration, stance duration, and swing phase portion displayed a significant difference between HC and LEF. These results show that the LEF group has a significantly slower walking pace than HC, similar to that of the IC group, indicating that this group can be considered at risk of cognitive impairment because a slowing of the gait was determined a predecessor to the cognitive deterioration [4, 6, 14]. On the other hand, longer loading response and double limb support phase, as well as the shorter single limb support phase present in IC, are all indicators of poor stability during gait, which is not the case with LEF.
During LR the bodyweight needs to be transferred onto a limb that just completed its forward swing [15]. To enable this, the vestibular system needs to accurately determine which limb position would result in retaining balance. It has been suggested that there is a phase-based vestibular information weighing, and according to Bent [16], vestibular information received at heel contact is held at higher importance when walking. One hypothesis is that a long time of gait cycle spent in LR indicates a slower adjustment of the vestibular system. Such would imply that the IC participants have an impairment of the vestibular system. The LR and PS phases are sub-phases of double limb support, so the increased percentage spent in LR, and PS is accompanied by the increase in double limb support percentage. On the other hand, the single limb support percentage can be regarded as a representative index of support capacity of the affected limb [15]. Therefore, a shorter percentage would indicate less support capacity, or in other words, less stability.
According to the gait function analysis results, participants in the IC group exhibit poorer balance than those in the HC and LEF groups. Postural stability testing confirmed this indication. There were significant differences between HC and IC for all postural stability indexes and there were significant differences between LEF and IC in the case of mediolateral and overall postural stability index. As suggested by earlier studies, this finding brings further confirmation that low postural stability is a sign of cognitive deterioration [17].
The results of pure tone audiometry revealed the highest average ability in LEF. However, apart from the sentence recognition test score and number of mistakes, the auditory variables showed no significant differences between the three groups. The LEF group was shown to have the greatest auditory retention ability. This ability is closely related to attention and perceptual processing, as is visual capacity. Similar to the auditory ability, even though the statistical analysis showed no significant differences between the groups, the LEF group was found to have the greatest visual capacity.
Based on the post-hoc power analysis it can be concluded that the statistical tests that were performed had enough power to detect large effect sizes. Therefore, the aforementioned variables presented large statistically significant differences between the respective groups.
Sensory, postural stability and gait functions of the three groups have shown that some variables have a high discriminative ability. The variables that were found to be good at discerning between groups and could therefore be used in detection systems were (1) cadence during level walking for dividing cognitively healthy older individuals from the rest and (2) single limb support portion, mediolateral stability index, and the number of mistakes on the sentence recognition test for discerning between older adults at risk of cognitive impairment from the ones with cognitive impairment.
The radial graph representing gait, balance, and sensory functions shows that the sensory ability is on average the highest in the LEF group, despite the gait and postural ability being the highest in the case of HC. Therefore, the question is: what does this say about the sensory and cognitive function association? Several hypotheses attempt to explain how the sensory systems and cognitive function are associated, one of which is the information degradation hypothesis. According to this hypothesis, when the sensory periphery is impaired, the degraded sensory input places an increased demand on the processing resources. These resources are considered to be limited in the amount of information that can be held in memory [18]. For example, when the quality of the auditory signal is degraded by environmental noise, or hearing loss, the ‘listening effort’ needed for processing and comprehending increases. This in turn diverts the limited cognitive resources towards effortful listening [19, 20], leaving no cognitive resources available for other tasks. It has also been suggested that the age-related cognitive changes stem from the age-related changes in sensory processing [21]. In a study done by Karawani [22], two groups of hearing-matched older adults, one that was given a hearing aid for the first time and one without it, were compared after a period of 6 months. At the end of the trial, enhanced working memory performance and increased cortical response were observed in the group with a hearing aid. These findings suggest that sensory restoration can free up available cognitive resources for remembering the spoken conversation. As our results showed, the number of mistakes participants made while recalling sentences was not correlated to the pure tone audiometry results. This could be due to sentence retention requiring more cognitive resources than word retention or listening for a pure tone. Hence, the effort put into listening may be enough for the cognitively impaired to hear the tones or words played for them just as well as the ones with normal cognition, but not enough to accurately remember whole sentences. In regard to vision, untreated poor vision was found to be a contributing factor to dementia in older individuals [23], and patients with dementia used less visual correction, had fewer ophthalmological treatments, and underwent fewer ocular surgeries [24]. Such findings indicate that treating the sensory periphery can aid in stopping or even reversing cognitive decline.
Limitations
One limitation of this study is the absence of longitudinal data which would follow the course of cognitive function of the participants. However, because in longitudinal studies a percentage of participants drops out, a larger sample is needed to reach meaningful conclusions. Therefore, another limitation is the size of the dataset, as a larger sample size would allow for more precise distinction between groups and even in designing a classification model. Additionally, higher resolution testing may reveal more about the visual ability of the three groups.