Table 1.
Study characteristics
| Study | Country | Model derivation, external validation or incremental value? | Outcome | Setting | Sample number; total (n=outcome) |
Validation | Predictors included | Model performance | How is the model presented in paper? |
| Antwi et al 44 | Ghana | Derivation | Gestational hypertension |
Lower-middle income | 2529 (n=261) | External | History of hypertensionin parents, GH in aprevious pregnancy, diastolic blood pressure (mm Hg), height (cm), weight (kg) and parity | C-statistic 0.70 (95% CI 0.67 to 0.74 | Final model is presented. Score chart is presented. |
| Antwi et al 23 | Ghana | Incremental value | Gestational hypertension | Lower-middle income | 373 (n=25) | Internal | Added predictors: PAPP-A and PlGF | AUC: 0.95 (95% CI 0.87 to 1.00) | |
| Benjamin et al 45 | India | Derivation | Cephalopelvic disproportion | Lower-middle income | 249 (n=27) | None | Foot length, intertrochanteric diameter and estimated fetal weight | Positive predictive value 34.0 | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
| de Oliveira et al 18 | Brazil | Derivation | Spontaneous preterm labour | Upper-middle income | 70 (n=23) | Internal | Age, skin colour, educational level, family income, parity, number of preterm deliveries, gestational age at the examination, cervical length and cervical glandular area | AUC: 0.91 (sensitivity 88%, specificity 93%) | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
| Geelhoed et al 46 | Ghana | Derivation | Anaemia in pregnancy | Lower-middle income | 327 (n=175) | None | Number of antenatal visits and lowest Hb (g/dL) | Explanation of 25% of the variability in the Hb prior to childbirth among women with severe anaemia | Final model is presented and how to incorporate the model is explained. |
| Harutyunyan et al 47 | Armenia | Derivation | Pre-eclampsia in multiparous women | Lower-middle income | 184 (n=36) | None | Interbirth interval, time to pregnancy, BMI, barrier methods of contraception and household monthly income | Hosmer-Lemeshow test: 5.0 (df=8, p>χ2=0.76) | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
| Hoirisch- Clapauch and Benchimol-Barbosa48 | Brazil | Derivation | Pre-eclampsia | Upper-middle income | 238 (n=79) | Internal | Oligomenorrhoea, weight accrual rate >0.8 kg/year, triglyceridemia and elevated acanthosis nigricans in the neck | C-statistic 0.81 (95% CI 0.77 to 0.84) (sensitivity 73%, specificity 85%) | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
| Kayode et al 49 | Nigeria | Derivation | Stillbirth | Lower-middle income | 6956 (n=443) | Internal | Maternal comorbidity, place of residence, maternal occupation, parity, bleeding in pregnancy and fetal presentation | C-statistic: 0.80 (95% CI 0.78 to 0.83) | Final model is presented. Score chart is not presented. |
| Kumar et al 50 | India | Derivation | Gestational hypertension | Lower-middle income | 2190 (n=198) | None | BMI, mean arterial pressure, early diastolic uterine artery notch on the left side, PAPP-A MoM, free β-hCG MoM | AUC: 0.810 (95% CI 0.746 to 0.875) | Final model is presented. Score chart is not presented. |
| Nascimento et al 40 | Brazil | Derivation | Neonatal death | Upper-middle income | 1351 (n=58) | Internal | Birth weight, gestational age, Apgar score and previous report of stillbirth | AUC: 0.90 (95% CI 0.84 to 0.96) | Final model with examples is presented. |
| Payne et al 19 | Fiji, Uganda, South Africa, Brazil and Pakistan | Derivation | Gestational hypertension-related complications | Upper-middle income, lower-middle income and low income | 2081 (n=261) | Internal and external | Multiparity, gestational age at admission, systolic blood pressure, 2+, 3+ and 4+ dipstick proteinuria, vaginal bleeding with abdominal pain, headache and/or visual changes and chest pain and/or dyspnoea | AUC: 0.768 (95% CI 0.735 to 0.801) | Final model is presented. Score chart is not presented. |
| Payne et al 20 | Fiji, Uganda, South Africa, Brazil and Pakistan | Derivation | Stillbirth and neonatal deaths | Upper-middle income, lower-middle income and low income | 1688 (n=110) | Internal | Maternal age, indicator for presence of one symptom, indicator for presence of two or more symptoms, indicator for dipstick proteinuria of 2+ or 3+, indicator for dipstick proteinuria of 4+ | AUC: 0.75 (95% CI 0.71 to 0.80) | Final model is presented. Score chart is not represented. |
| Payne et al 21 | Fiji, Uganda, South Africa, Brazil and Pakistan | Incremental value | Gestational hypertension-related complications | Upper-middle income, lower-middle income and low income | 852 (n=119) | None | Added predictors: SpO2 |
AUC: 0.71 (95% CI 0.66 to 0.77) | Final model represented Score chart is not presented. |
| Phaloprakarn et al 51 | Thailand | Derivation | Gestational diabetes | Upper-middle income | 1876 (n=586) | External | Age, BMI, family history of diabetes, prior macrosomia and history of >2 abortions | AUC: 0.77 (95% CI 0.75 to 0.79) | Score chart is presented. |
| Prata et al 52 | Egypt | Derivation | Postpartum haemorrhage | Lower-middle income | 2510 (n=93) | None | Elevated number of antenatal care visits, history of postpartum haemorrhage, anaemia, labour augmentation, retained placenta, length of first and second stage | 4+ risk factors: sensitivity 14.79% and specificity 98.92%. Likelihood ratio 12.76 | Final model is not presented. Number of risk factors with prognostic values presented. |
| Romero-Gutiérrez et al 53 | Mexico | Derivation | Stillbirth | Upper-middle income | 753 (n=753) | None | Maternal age, antenatal care and umbilical cord complication | R2: 0.28 (p<0.001) | Final model is presented. Score chart is not presented. |
| Sekizawa et al 54 | Indonesia | Derivation | Pre-eclampsia | Lower-middle income | 660 (n=62) | None | Endoglin, vascular endothelial growth factor receptor-1, placental growth factor and parity | AUC: 0.88 (95% CI 0.84 to 0.92) | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
| Tsu55 | Zimbabwe | Derivation | Postpartum haemorrhage | Low income | 653 (n=151) | None | Age 35+ years, low parity (0, 1), poor obstetric outcome last pregnancy, antenatal Hb <12 g/dL, antenatal hospitalisation for pregnancy-related problem | Sensitivity 80%, sensitivity 50% | Final score system presented. |
| Ukah et al 22 | Fiji, Uganda, South Africa, Brazil and Pakistan | External validation | Gestational hypertension-related complications | Upper-middle income, lower-middle income and low income | 2081 (n=261) | External | Gestational age, chest pain or dyspnoea, oxygen saturation, platelet count, serum creatinine and serum aspartate aminotransferase | AUC: 0.77 (95% CI 0.72 to 0.82) | Final model is presented. Score chart is not presented. |
| Zhou et al 56 | China | Derivation | Pre-eclampsia | Upper-middle income | 1000 (n=61) | None | HDLc, uric acid and triglycerides | AUC: 0.77 (sensitivity 92%, specificity 50%) | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
| Zhou et al 56 | China | Derivation | Gestational diabetes | Upper-middle income | 1000 (n=100) | None | HDLc, uric acid and triglycerides | AUC: 0.71 (sensitivity 80%, specificity 53%) | Final model predictors are presented. Score chart or final model with assigned weights is not presented. |
AUC, area under the curve; BMI, body mass index; GH, Gestational Hypertension; Hb, haemoglobin; hCG, human chorionic gonadotropin; HDL, high-density lipoprotein; MoM, multiples of the normal median; PAPP-A, pregnancy-associated plasma protein A; PlGF, placental growth factor.