Table 3.
Binary logistic regression model of predictors for PrE/E in PCMH
| Characteristics | Case n (%) | Control n (%) | COR (95% CI) | p-value | AOR (95%CI) | p-value |
|---|---|---|---|---|---|---|
| Maternal assets | ||||||
| ≤ 50,000 SLL | 82 (40.0) | 87 (19.7) | 2.72 (1.89–3.92) | < 0.001 | 2.56 (1.63–4.02) | < |
| > 50,000 SLL | 123 (60.0) | 355 (80.3) | 1.0 | 1.0 | 0.001 | |
| Family predisposition for preeclampsia and eclampsia | ||||||
| Yes | 23 (16.7) | 14 (3.6) | 5.33 (2.66–10.69) | < 0.001 | 2.72 (1.46–5.07) | 0.002 |
| No | 115 (83.3) | 373 (96.4) | 1.0 | 1.0 | ||
| Preexisting hypertension | ||||||
| Yes | 15 (7.2) | 9 (2.0) | 3.88 (1.67–9.01) | 0.001 | 3.64 (1.32–10.06) | 0.013 |
| No | 193 (92.8) | 449 (98.0) | 1.0 | 1.0 | ||
| UTI | ||||||
| Yes | 72 (33.6) | 73 (16.0) | 2.66 (1.82–3.88) | < 0.001 | 2.02 (1.28–3.19) | 0.002 |
| No | 142 (66.4) | 383 (84.0) | 1.0 | 1.0 | ||
| Prolonged diarrhoea | ||||||
| Yes | 72 (33.6) | 55 (12.0) | 3.71 (2.49–5.53) | < 0.001 | 2.81 (1.63–4.86) | < 0.001 |
| No | 142 (66.4) | 402 (88.0) | 1.0 | 1.0 | ||
| Malaria | ||||||
| Yes | 153 (71.5) | 279 (61.1) | 1.60 (1.13–2.27) | 0.008 | 1.43 (0.95–2.15) | 0.090 |
| No | 61 (28.5) | 178 (38.9) | 1.0 | 1.0 | ||
| MUAC | ||||||
| > 32 cm | 50 (23.9) | 48 (10.5) | 2.67 (1.73–4.14) | < 0.001 | 3.09 (1.83–5.22) | < 0.001 |
| ≤ 32 cm | 159 (76.1) | 408 (89.5) | 1.0 | 1.0 | ||
| Living close to a waste deposit | ||||||
| Yes | 59 (29.4) | 70 (16.3) | 2.14 (1.44–3.18) | < 0.001 | 1.94 (1.15–3.25) | 0.013 |
| No | 142 (70.6) | 360 (83.7) | 1.0 | 1.0 | ||
| Well or borehole water as main source of drinking water | ||||||
| Yes | 61 (28.9) | 80 (17.6) | 1.90 (1.30–2.79) | 0.001 | 2.05 (1.31–3.23) | 0.002 |
| No | 150 (71.1) | 374 (82.4) | 1.0 | 1.0 | ||
| Average sleep duration | ||||||
| Short < 6 h | 139 (70.2) | 247 (54.9) | 1.94 (1,36–2.77) | < 0.001 | 1.48 (0.96–2.29) | 0.078 |
| Long ≥6 h | 59 (29.8) | 203 (45.1) | 1.0 | 1.0 | ||
| Fruit intake | ||||||
| Inadequate | 75 (35.9) | 66 (14.5) | 3.30 (2.25–4.85) | < 0.001 | 2.58 (1.64–4.06) | < 0.001 |
| Adequate | 134 (64.1) | 389 (85.5) | 1.0 | 1.0 | ||
Binary Logistic Regression Model was calculated including variables matching criteria of a p-value ≤0.001 and crude odds ratios > 1.5. We additionally included the risk factor Malaria during pregnancy because of its previously described influence of Plasmodium falciparum infected placenta on PrE/E. Multiple Imputation was conducted with the Mersenne Twister as random number generator program and for generator initialization we chose the fixed value of 2,000,000. We imputed 25 variables with 10 datasets. Statistical significance was set at p-value < 0.05, crude and adjusted odds ratios and 95% confidence interval were calculated. The Hosmer Lemeshow goodness of fit test for logistic regression was applied and showed overall a good fit with a p-value > 0.05