Table 1.
Outcomes of rapid diagnostic testing in hypothetical cohorts of 100 people presenting with fever in settings with different malaria prevalences*
| Pretest probability of positive result (%) | No of cases/100 people (95% CI) | |||||
|---|---|---|---|---|---|---|
| Non-malarial cause of fever | Malaria is true cause of fever | |||||
| False positives† | True negatives‡ | True positives§ | False negatives¶ | |||
| 5 | 5 (3 to 6) | 90 (89 to 92) | 5 (5 to 5) | 0 (0 to 0) | ||
| 30 | 4 (3 to 5) | 66 (65 to 67) | 29 (28 to 29) | 2 (1 to 2) | ||
| 60 | 2 (2 to 3) | 38 (37 to 38) | 57 (56 to 58) | 3 (2 to 4) | ||
| 80 | 1 (1 to 1) | 19 (19 to 19) | 76 (75 to 77) | 5 (3 to 5) | ||
*Type 1 HRP-2 rapid diagnostic test (RDT) with average sensitivity of 94.8% (95% CI 93.1 to 96.1) and specificity 95.2% (95% CI 93.2 to 96.7).12
†Number of unnecessary prescriptions that would still occur when using RDTs (the true cause of fever may also go untreated).
‡Number of unnecessary antimalarial prescriptions that could be avoided if RDTs are used instead of presumptive treatment.
§Number of people correctly diagnosed with malaria by the RDT.
¶Number of people with malaria who would be sent home without antimalarials because of a negative RDT result.