Table 2.
Multiple logistic regression models showing the association between the highest tertile of stride-to-stride variability of stride time (dependent variable) and Digit Span score (independent variable) adjusted for clinical characteristics (n = 78)
| Full adjusted model | Stepwise backward model | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Digit Span score * | 0.78 | [0.63;0.96] | 0.020 | 0.81 | [0.68;0.97] | 0.019 |
| Age | 1.02 | [0.50;2.10] | 0.954 | - | - | - |
| Female | 1.05 | [0.32;3.4] | 0.932 | - | - | - |
| BMI | 1.09 | [0.94;1.27] | 0.253 | - | - | - |
| Education level † | 1.11 | [0.95;1.30] | 0.182 | - | - | - |
| Number of drugs taken daily | 1.00 | [0.77;1.30] | 0.984 | - | - | - |
| History of falls ‡ | 2.07 | [0.57;7.54] | 0.270 | - | - | - |
OR: Odds ratio
CI: confident interval
BMI: body mass index
*: Total number of digits that a subject can absorb and recall in correct forward and backward serial orders after hearing them
†: Assessed with the number of years at school
‡: Falls in the previous year
Odds ratio significant (i.e.; < 0.05) indicated in bold.