Table 2.
Comparison of three modelsa in terms of their covariate effects on the presence of concurrent depression (having five or more depressive symptoms)
| MIXED MODEL |
WEIBULL MODEL |
DISCRETE TIME MODEL |
||||
|---|---|---|---|---|---|---|
| COVARIATES | ESTIMATED OR (95% CI) | P VALUE | ESTIMATED OR (95% CI) | P VALUE | ESTIMATED OR (95% CI) | P VALUE |
| Age 75–84 years | 1.13 (0.80, 1.60) | 0.49 | 1.04 (0.51, 2.10) | 0.92 | 1.04 (0.51, 2.10) | 0.92 |
| Age ≥85 years | 0.88 (0.40, 1.97) | 0.76 | 0.63 (0.13, 3.05) | 0.57 | 0.63 (0.13, 3.08) | 0.56 |
| Female gender | 1.80 (1.27, 2.56) | 0.001 | 2.05 (1.03, 4.08) | 0.04 | 2.05 (1.03, 4.09) | 0.04 |
| At least high school education | 0.43 (0.31, 0.59) | <0.001 | 0.33 (0.17, 0.63) | <0.001 | 0.33 (0.17, 0.63) | 0.001 |
| Use of ≥4 prescription drugs | 2.13 (1.48, 3.08) | <0.001 | 3.51 (1.65, 7.45) | 0.001 | 3.50 (1.64, 7.44) | 0.001 |
| Use of an antidepressant drug | 2.93 (1.38, 6.18) | 0.005 | 16.03 (2.36, 108.80) | 0.005 | 16.15 (2.35, 110.87) | 0.005 |
| Functional impairment | 2.75 (1.96, 3.87) | <0.001 | 4.73 (2.41, 9.31) | <0.001 | 4.72 (2.40, 9.31) | <0.001 |
| Time = wave 3† | 0.74 (0.54, 1.02) | 0.06 | 0.61 (0.39, 0.95) | 0.03 | 0.61 (0.39, 0.95) | 0.03 |
| Time = wave 4† | 0.69 (0.49, 0.98) | 0.04 | 0.58 (0.36, 0.93) | 0.02 | 0.57 (0.36, 0.93) | 0.02 |
| Time = wave 5† | 0.61 (0.41, 0.90) | 0.01 | 0.51 (0.30, 0.87) | 0.01 | 0.51 (0.30, 0.86) | 0.01 |
| Time = wave 6† | 0.75 (0.50, 1.14) | 0.18 | 0.73 (0.42, 1.29) | 0.28 | 0.73 (0.42, 1.29) | 0.28 |
| Variance of the shared random effect | N/A | N/A | 18.70 (7.46, 29.94) | 0.001 | 18.94 (7.56, 30.33) | 0.001 |
| CPU (min:sec) | 1.03 | 15:19.10 | 31:03.37 | |||
| AIC | N/A | 3599.7 | 3453.9 | |||
a
All models are multivariable models with intercept (estimate not shown).
†
Effect of wave at which data were collected (with wave 2 as the reference group).
OR = odds ratio; CI = confidence interval; CPU = central processing unit (shorter is computationally faster); AIC = Akaike information criteria (smaller indicates a better fit); N/A = not applicable.