Table 4.
Sensitivity analysis after excluding GDM patients
| Quartiles of energy-adjusted carotenoid intakes (OR, 95% CI) | Ptrend a | Per one-SD increase b | ||||
|---|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | |||
| Total carotenoids | ||||||
| Crude | 1.0 | 0.75 (0.51, 1.09) | 0.61 (0.41, 0.90)* | 0.49 (0.32, 0.75)** | 0.001 | 0.71 (0.60, 0.84)** |
| Adjusted model | 1.0 | 0.57 (0.31, 1.03) | 0.47 (0.24, 0.90)* | 0.36 (0.18, 0.69)** | 0.002 | 0.65 (0.51, 0.84)** |
| α-carotene | ||||||
| Crude | 1.0 | 1.09 (0.74, 1.60) | 0.93 (0.63, 1.37) | 0.71 (0.46, 1.09) | 0.06 | 0.91 (0.79, 1.05) |
| Adjusted model | 1.0 | 0.89 (0.49, 1.59) | 0.87 (0.47, 1.62) | 0.67 (0.35, 1.30) | 0.24 | 0.94 (0.77, 1.15) |
| β-carotene | ||||||
| Crude | 1.0 | 0.72 (0.49, 1.06) | 0.63 (0.42, 0.94)* | 0.55 (0.36, 0.83)* | 0.004 | 0.78 (0.67, 0.91)* |
| Adjusted model | 1.0 | 0.61 (0.33, 1.11) | 0.37 (0.19, 0.71)* | 0.36 (0.18, 0.70)* | 0.001 | 0.76 (0.61, 0.95)* |
| β-cryptoxanthin | ||||||
| Crude | 1.0 | 0.92 (0.63, 1.34) | 0.93 (0.61, 1.41) | 0.75 (0.49, 1.14) | 0.18 | 0.73 (0.59, 0.91)* |
| Adjusted model | 1.0 | 0.97 (0.54,1.74) | 0.79 (0.41, 1.54) | 0.49 (0.26, 0.94)* | 0.02 | 0.68 (0.48, 0.98)* |
| Lycopene | ||||||
| Crude | 1.0 | 1.01 (0.68, 1.51) | 0.81 (0.54, 1.21) | 0.59 (0.39, 0.90)* | 0.006 | 0.69 (0.57, 0.84)** |
| Adjusted model | 1.0 | 0.69 (0.38, 1.25) | 0.58 (0.30, 1.10) | 0.61 (0.32, 1.14) | 0.14 | 0.70 (0.53, 0.93)* |
| Lut-zea | ||||||
| Crude | 1.0 | 0.81 (0.56, 1.18) | 0.62 (0.40, 0.94)* | 0.59 (0.39, 0.91)* | 0.01 | 0.82 (0.69, 0.96)* |
| Adjusted model | 1.0 | 0.59 (0.31, 1.10) | 0.49 (0.25, 0.95)* | 0.35 (0.18, 0.69)* | 0.003 | 0.67 (0.51, 0.87)* |
GDM Gestational diabetes mellitus, OR Odds ratio, CI Confidence interval, Q Quartile, SD Standard deviation, Lut-zea Lutein and zeaxanthin
a Tested by entering the median intake of each quartile of carotenoids as a continuous variable into univariate and multivariate logistic regression models. b Performed by standardising the energy-adjusted carotenoid intakes data and entering it into the model.
Adjusted model was adjusted for age (years), gestational age (weeks), pre-pregnancy BMI (kg/m2), gravidity, parity, gestational diabetes mellitus (GDM) status (yes/no), family hypertension history (yes/no), physical activity (MET [h/d]), education and income levels and total energy intake (kcal/d)
*P < 0.05; **P < 0.001.