Table 6. EPIFIL predictions of timelines to achieve 1% mf prevalence in Dokan Tofa, Nigeria.

Site	Scenario#	No. of accepted models	Median no. of years (2.5th-97.5th percentiles) (significance*)	Weighted variance (significance*)	Entropy (significance*)	Relative information gained by data (%)+
Dokan Tofa	0	3007	3 (2–10)	2.411,2,3,4	2.551,2,3,4	-
	1	2046	3 (2–8)	2.400,2,3	2.450,2,3	0.41
	2	2007	3 (2–7)	2.070,1,4	2.350,1,4	0.85
	3	2007	3 (2–7)	2.070,1,4	2.350,1,4	0.85
	4	2046	3 (2–8)	2.400,2,3	2.450,2,3	0.41

The lowest entropy scenario for each site is bolded and shaded grey. Additional scenarios shaded grey are not significantly different from the lowest entropy scenario.

^#Scenario 0: model-only; Scenario 1: baseline data; Scenario 2: baseline + post-MDA 3 data; Scenario 3: baseline + post-MDA 3 + post-MDA 5 data; Scenario 4: baseline + post-MDA 5 data

*For each pair of scenarios, pairwise F-tests for equality of variance were performed to compare the weighted variance, differential Shannon entropy tests were performed to compare the entropy, and Kruskal-Wallis multiple comparison tests were performed to compare medians. Pairwise significance is represented by reporting those scenarios which are statistically significantly different from each other by numbers (0–4) as superscripts. For example, the weighted variance for scenario 0 for Kirare has the superscript numbers (1–4) to indicate that the weighted variance for scenario 0 is significantly different from the weighted variance for scenarios 1–4. Significance was determined using the Benjamini-Hochberg procedure for controlling the false discovery rate (q = 0.05) in all pairwise statistical tests.

⁺information gained by each data stream (scenario 1–4) are presented in comparison to the information contained in the model-only simulation (scenario 0)