N | Study | Year | Period of participant recruitment | Country | Type of study | Sample size | Statistics used | Name of model | Method internal of validation | Age profiles | Sex profiles | Outcome predicted | Variables used | Diagnostic properties | External validation | Use in clinical settings |
Mortality | ||||||||||||||||
1 | Jaffar et al 21 | 1997 | 1992–1994 | Gambia | Retrospective analysis of data from a randomised control trial | 624 | Logistic regression | None | None | 1–9.5 years | Females—49% | Mortality in paediatric cerebral malaria | Cold periphery, deep coma and hypoglycaemia | Not done | None | NE |
2 | Molyneux et al 27 | 1989 | January 1987 to June 1988 | Malawi | Cohort | 131 | Univariable analysis | Bedside prognostic index | None | 7 months to 10 years | Females—55.7% | Mortality in paediatric cerebral malaria | Blood glucose, parasitaemia, WBC count, age, coma score, absent corneal reflexes, decerebration, convulsions | Positive predictive value—83%, sensitivity—66% | None | NE |
3 | Conroy et al 16 | 2012 | 1997–2009 | Malawi | Cohort | 155 | Logistic regression | None | Hosmer-Lemeshow goodness-of-fit test | 8 months to 14 years | Females—54.4% | Mortality in patients with cerebral malaria | Age, Blantyre coma score, respiratory distress, severe anaemia, angiopoietin-1, angiopoietin-2 and sTie-2 levels | C-index of 0.79 (95% CI 0.72 to 0.84) | None | NE |
4 | Krishna et al 22 | 1994 | 1988–1989 | Gambia | Cohort study | 115 | Logistic regression | None | Wald statistic and ROC analysis | 18 months to 12 years | NC | Mortality in paediatric severe malaria | Coma score, whole blood lactate/glucose ratio, TNF level | Wald statistic: coma score (4.5), lactate/glucose ratio (8.36), TNF level (6.5) | None | NE |
5 | Marsh et al 23 | 1995 | May 1989 to November 1991 | Kenya | Cohort | 1844 | Logistic regression | None | None | Mean: 26 months | NC | Mortality in children with severe malaria | Impaired consciousness, respiratory distress, hypoglycemia, and jaundice | Predicted 92.2% of deaths | None | NE |
6 | Newton et al 28 | 2005 | January 2001 to December 2003 | Malawi, Kenya and Ghana | Cohort | 14 605 | Linear regression | None | AUROC | Mean age: 32–36 months | Females—53%–55% | Mortality in paediatric severe falciparum malaria | Deep breathing, Blantyre Coma Score, inability to sit, weight-for-age Z score, hypoglycaemia, base excess and lactate concentration | c-statistic 0.83–0.88 in the three sites: Blantyre (0.88), Kilifi (0.87) and Kumasi (0.83) | None | NE |
7 | Gérardin et al 32 | 2006 | 1 October 1997 to 31 March 1999 | Senegal | Cohort | 311 | Logistic regression | PRISM (paediatric risk of mortality) AUC: 0.9240 | Hosmer-Lemeshow χ2 test | Median: 8 years (IQR: 5–11 years) | Females—40.5% | Mortality in children with falciparum malaria | Systolic blood pressure, temperature, mental status, heart rate, dilatation of pupils, pH, total CO2, PCO2, arterial PaO2, serum glucose, potassium, urea, creatinine, white blood cells, prothrombin time, platelet count | AUROC for acute malaria: 0.89 (95% CI 0.85 to 0.92) and 0.86 (95% CI 0.81 to 0.90) for severe malaria | Yes | NE |
8 | Helbok et al 20 | 2009 | December 2000 to May 2005 | Gambia, Malawi, Kenya, Ghana, and Gabon | Cohort | 23 890 | Logistic regression | LODS (Lambaréné Organ Dysfunction Score) | Internal validation using Bonferroni correction | Mean: 30–38 months | Females—41%–47% | Mortality in children with severe falciparum malaria | Coma, prostration and deep breathing | AUROC: 0.80 (0.79 to 0.82) | Yes | NE |
9 | von Seidlein et al 30 | 2012 | 2005–2010 | Gambia, Mozambique, Nigeria, Rwanda, Kenya, DRC, Tanzania, Ghana, Uganda | Retrospective analysis | 5426 | Logistic regression | None | ROC analysis | Median: 2.8 years (1.7, 4.3) | NC | Mortality in paediatric severe falciparum malaria | Base deficit, coma, convulsions, BUN and chronic illness | AUROC: 0.85 (95% CI 0.83 to 0.87) | None | NE |
10 | Conroy et al 31 | 2015 | NC | Uganda | Cohort | 1589 | Logistic regression | SICK (Signs of Inflammation in Children that Kill) 38—AUC*: 0.887 (sensitivity 84.1% specificity 82.2%) | Hosmer-Lemeshow goodness-of-fit | NC | Females—54.3% | Mortality in malaria | Altered consciousness, temperature, heart rate, respiratory rate, systolic blood pressure, capillary refill time and age | AUROC: 0.846 | Yes | NE |
LODS57 | Hosmer-Lemeshow goodness-of- fit | NC | Females—54.3% | Mortality in malaria | Prostration, coma (BCS) and deep breathing | AUROC: 0.898 | Yes | NE | ||||||||
PEDIA39—AUC*: 0.93 (95% CI 0.92 to 0.94) | Hosmer-Lemeshow goodness-of- fit | NC | Females—54.3% | Mortality in malaria | Kwashiokor†, jaundice, subcostal indrawing, prostration (±seizures) and wasting | AUROC: 0.896 | Yes | NE |
*Diagnostic properties of original model.
†Not used in present model.
AUROC, area under the receiver operating curve; BCS, Blantyre coma score; BUN, blood urea nitrogen; DRC, Democratic Republic of the Congo; NC, not clear; NE, no evidence; TNF, tissue necrotic factor; WBC, white blood cells.