. 2010 Sep 21;10:275. doi: 10.1186/1472-6963-10-275

Table 4.

The effects of interventions on diagnostic resolution, diagnostic accuracy of dengue and prescription of antibiotics

Interventions	Intervention effects, M-H Odds ratios of association between pre/post intervention and correctness of diagnosis (95% CI)

	Diagnostic resolution		Diagnostic Accuracy of Dengue ³⁾	Prescription of Antibiotics

	"Undifferentiated fever"¹⁾	"Dengue" ²⁾
Training	2.1 (1.6-2.8)^a)	0. 3 (0.2-0.5)^a)	0.8 (0.3-2.0) ^c)	1. 1 (0.7-1.5)^c)
RDTs	1.3 (1.0-1.8) ^b)	0.3 (0.1-0.6) ^a)d)	1. 4 (0.8-2.4)^c)	0. 7 (0.5-0.9)^a)
Training + RDTs	0.9 (0.7-1.1)^c)	1.5 (1.0-2.2)^c)	1.9 (1.1-3. 5)^b)d)	0.1 (0.0-0.1)^a)
Control	1.2 (0.9-1.6)^c)d)	0.3 (0.2-0.5) ^a)d)	1.1 (0.4-3.2)^c)	10.3 (6.9-15.4)^a)

1) "undifferentiated fever" as a presumptive diagnosis, compared to all more specific diagnoses; 2) "dengue" as a presumptive diagnosis, compared to diagnoses not being dengue; 3) the serologically confirmed presumptive diagnosis of dengue or not dengue.

a) p-value: ≤ 0.002; b) 0.05 > p-value ≥ 0.01; c) p-value: ≥ 0.05 d) significant homogeneity among individual health staff members within the intervention group.