When investigating age-related effects on functional mobility, a critical controversy arises relating to fundamental differences in the definition of the term "normal ageing". On the one hand, normal older people can be defined as only those free from all medical conditions, whilst on the other end, all older people, with no exclusion criteria and hence representative of the general population, can be considered normal. While both perspectives on selection criteria are valid, they lead to differing results, depending on whether pathological conditions are considered as a normal concomitant of the ageing process. The older sample on whom the data analysis was conducted was representative of the community-living older population and thus presented with a range of pathologies.
The study findings revealed significant age-related differences in all seven functional mobility tests examined. These findings confirm those of previous studies and indicate that when compared with young people, older people exhibit poorer leaning balance [1, 2], more difficulty maintaining balance while standing with a reduced base of support , slower comfortable walking speed [6, 7, 10], reduced ability to quickly rise from a chair , and slower stair ascent and descent speed . These age-related differences in functional mobility have been attributed to impaired sensorimotor function [19, 20], in particular reduced lower extremity strength and power [19–22], but also to balance deficits [19, 20], increased fear of falling [20, 23] and reduced aerobic capacity .
Significant correlations among all the functional mobility tests in the older group indicate that older adults who performed poorly in one test were likely to perform poorly in all the other tests. This suggests that to a large extent these tests assess a common underlying "mobility" construct , rather than distinct functional abilities.
The finding that the older women performed worse than the older men in all the functional mobility tasks is in agreement with previous studies that have investigated lower-extremity functional performance , stair negotiation , rapid turns and stops , and is attributed to older women being less able to generate rapid lower limb muscle torques [20, 26].
The tests differed considerably with regard to differences in performances between the young and older groups. The six metre walk test showed the smallest age difference and this is likely due to the test instruction requiring walking at normal rather than maximal pace, and the fact that this test is familiar and of low threat with respect to falling. In contrast, the stair descent test, which required participants to undertake the task as quickly as possible, is likely to have induced the greatest concern about falling and this was evident in different strategies adopted by the young and older participants. Only one young participant held the handrail while negotiating the stairs and many "ran" rather than walked down the stairs, while approximately half of the older people held the handrail and none adopted a running strategy. As 28% of the older sample reported moderate or marked fear of falling, this factor, in addition to sensorimotor function impairments, may have contributed to the large difference in stair descent speed between the young and older groups.
The greatest age-related differences in test performance were found in the coordinated stability and near tandem balance tests. These tests were completed without error by most young participants, but proved to be much more difficult for the older participants. This suggests that the ability to control and adjust standing balance may undergo greater age-related changes than transfer and walking tasks. However, it is also possible that the larger age effects may be partly due to familiarisation factors in that the coordinated stability and near tandem balance tests are less similar to everyday tasks than tests such as the sit to stand and stair negotiation which are integral elements of activities of daily living.
Normative data regarding functional mobility performance in older people suffering from two common medical conditions in our sample, stroke and arthritis, were also provided. As suggested in previous studies, sensory and motor control impairments likely contributed to reduced functional abilities in stroke survivors  and arthritis sufferers . Surprisingly though, the difference in functional tests performance was not as large between stroke sufferers and non-stroke sufferers as it was between arthritis sufferers and non-arthritis sufferers. We did not assess the extent of damage and subsequent recovery from the stroke; it is likely that some of the older participants had functionally recovered from their stroke event which would explain the great variance in the scores. In contrast, the presence of arthritis would have been affecting the participants' mobility and balance on a daily basis.