Table 3. Independent effects of predictors for new GL occurrence*.
| Risk factors for GL | GL incidence | GL recurrence | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p1 | OR | 95% CI | p2 | |
| Cognitive function (per point increment) a | ||||||
| MMSE | 1.17 | .920~1.50 | .196 | 1.01 | .833~1.23 | .908 |
| Remote memory | 1.03 | .700~1.51 | .895 | .968 | .708~1.33 | .836 |
| Orientation | .790 | .650~.961 | .018 | .990 | .851~1.15 | .895 |
| Abstract thinking | .746 | .482~1.15 | .187 | .952 | .668~1.36 | .788 |
| Verbal fluency | .817 | .614~1.09 | .163 | 1.09 | .809~1.46 | .580 |
| Behavioral manifestation (per point increment) a | ||||||
| Landmark and scene agnosia | 1.41 | .788~1.02 | .159 | .873 | .680~1.12 | .288 |
| Egocentric disorientation | 1.50 | .860~2.61 | .153 | 1.29 | .894~1.86 | .147 |
| Inattention | 2.13 | 1.28~3.56 | .004 | .964 | .731~1.27 | .793 |
| Heading disorientation | .700 | .416~1.18 | .181 | 1.01 | .793~1.29 | .937 |
| Self-awareness a | ||||||
| No safety range vs. with safety range | .867 | .297~2.53 | .749 | 3.80 | 1.14~12.66 | .029 |
| Discrepancy score (per point increment) | 1.13 | 1.02~1.24 | .015 | .972 | .909~1.04 | .401 |
| Composite analysis a, b | ||||||
| Orientation | .830 | .714~.964 | .015 | .974 | .875~1.08 | .633 |
| Inattention | 1.85 | 1.08~3.16 | .026 | .983 | .740~1.31 | .907 |
| No safety range | 1.03 | .313~3.39 | .960 | 4.26 | 1.27~14.29 | .019 |
| Discrepancy score | 1.06 | .928~1.22 | .379 | .969 | .899~1.04 | .402 |
Abbreviations: GL = getting lost; OR = odds ratio; p1 = p value within Group A; p2 = p value within Group B.
a Adjusted for age, sex, and years of education.
b Including only previous significant measures.
*Values were determined by putting the predictors into each model (or cluster) simultaneously.