Assessing the interruption of the transmission of human helminths with mass drug administration alone: optimizing the design of cluster randomized trials

Table 5 Confidence limits (95%) for PPV and NPV values for simulations presented in Fig. 7. Prevalence threshold 30.0% three months after cessation of MDA. Data from 300 village simulations with 500 people per village

Raw data from 300 simulations	Bounce back	Transmission interruption	Totals
Test positive	15	254	269
Test negative	21	10	31
Totals	36	264	300
		95% Confidence limits
	Estimated value	Lower limit	Upper limit
Prevalence of interruption	0.880	0.836	0.913
Sensitivity	0.992	0.929	0.980
Specificity	0.583	0.409	0.740
Probability of simulation being either transmission interruption or bounce back
Interruption	0.9	0.855	0.930
Bounce back	0.1	0.072	0.145
For predicted elimination - probability of being true or false
True positive (PPV)	0.944	0.908	0.967
False positive	0.055	0.032	0.092
For predicted bounce back - probability of true or false
True negative (NPV)	0.677	0.485	0.826
False negative	0.322	0.173	0.514
Likelihood ratios
Positive (C)	2.309	1.568	3.401
Negative (C)	0.064	0.034	0.124
Positive (W)	16.933	10.347	27.713
Negative (W)	0.476	0.277	0.817

ISSN: 1756-3305