Table 2.
Fixed Effects for Multilevel Models of Change in Fast Walking Speed Before Death or End of Follow-up
| Fixed Effects | Model 1 | Model 2* | ||
|---|---|---|---|---|
| Estimate (95% CI) | p | Estimate (95% CI) | p | |
| Intercept† | 1.551 (1.537, 1.567) | <.0001 | 1.581 (1.560, 1.603) | <.0001 |
| Sex (women vs men) | −0.241 (−0.257, −0.226) | <.0001 | −0.192 (−0.215, −0.170) | <.0001 |
| Age (centered at 80 y), per 10-y increment) | −0.152 (−0.174, −0.131) | <.0001 | −0.152 (−0.174, −0.131) | <.0001 |
| Vital status (dead vs alive) | −0.090 (−0.113, −0.067) | <.0001 | −0.074 (−0.097, −0.051) | <.0001 |
| Time (annual change) | −0.021 (−0.022, −0.020) | <.0001 | −0.020 (−0.022, −0.018) | <.0001 |
| Time × age | −0.006 (−0.009, −0.002) | .002 | −0.006 (−0.010, −0.003) | .0004 |
| Time × vital status | −0.009 (−0.013, −0.005) | <.0001 | −0.009 (−0.014, −0.005) | <.0001 |
Notes: CI = confidence interval. *Adjusted for sex, age, vital status, education, height, smoking ≥20 pack-years, homocysteine, time, time × age, time × vital status, time × education, time × height, time × smoking ≥20 pack-years, time × homocysteine, time-dependent variables (body mass index, physical activity, coronary/peripheral artery disease, antihypertensive drugs, diabetes, lipid-lowering drugs, mini-mental state examination score, psychotropic drugs).
†In model 1, the intercept corresponds to the average walking speed at the end of follow-up for men alive and aged 80 y. In model 2, the intercept corresponds to the average walking speed at the end of follow-up for men alive and aged 80 y, with primary school level, body mass index = 25 kg/m2, height = 165 cm, physically active, smoking <20 pack-years, no health conditions, mini-mental state examination = 30, homocysteine = 14 µmol/L.