Table 2.
Model-fitting result of the prospective analysis of depression as a predictor to glycemic control, controlling for covariates (fixed effects) and adjusted for within-person change over time (random effect), number of observations = 577.
| HbA1c | ||||||
|---|---|---|---|---|---|---|
| Estimate | Std. Error | 95% CI | Wald’s p | |||
| upper | lower | |||||
| Intercept | 8.81 | 0.97 | 6.95 | 10.59 | <0.001 | |
| TGDS | −0.15 | 0.06 | −0.26 | −0.03 | 0.01 | |
| TGDS quadratic polynomial | 0.015 | 0.006 | 0.003 | 0.03 | 0.01 | |
| Age | −0.02 | 0.01 | −0.05 | 0.005 | 0.13 | |
| Female | 0.22 | 0.23 | −0.20 | 0.66 | 0.31 | |
| Duration of DM | 0.05 | 0.01 | 0.02 | 0.07 | <0.001 | |
| EQ VAS | −0.01 | 0.003 | −0.02 | −0.005 | <0.001 | |
| LDL cholesterol | 0.003 | 0.003 | 0.0001 | 0.006 | 0.04 | |
| Insulin used | 0.36 | 0.14 | 0.08 | 0.68 | 0.01 | |
| Risk of malnutrition | 0.17 | 0.16 | −0.14 | 0.48 | 0.29 | |
| Level of physical activity | Low | - | - | - | - | - |
| Moderate | 0.05 | 0.10 | −0.14 | 0.24 | 0.59 | |
| Vigorous | −0.06 | 0.13 | −0.30 | 0.20 | 0.65 | |
| Cigarette smoke | Never | - | - | - | - | - |
| Former | 0.22 | 0.26 | −0.27 | 0.70 | 0.24 | |
| Current | −0.35 | 0.27 | −0.86 | 0.17 | 0.18 | |
| Present of DR | −0.11 | 0.17 | −0.43 | 0.23 | 0.54 | |
| Present of DN | 0.19 | 0.13 | −0.06 | 0.43 | 0.13 | |
| BMI > 25 | −0.10 | 0.14 | −0.37 | 0.16 | 0.46 | |
| Random effects | Estimate | Std.Dev. | ||||
| Intercept | 0.65 | 0.80 | ||||
| Residual | 0.67 | 0.82 | ||||
TGDS = Thai Geriatric Depression Score, EQ VAS = EQ visual analogue, LDL= low-density lipoprotein, DR = diabetic retinopathy, DN = diabetic nephropathy, BMI = Body Mass Index, Std.Dev. = standard deviation.