Skip to main content
. 2017 Mar;18:67–80. doi: 10.1016/j.epidem.2017.01.002

Table 4.

Estimated parameter values resulting from fitting to all district data without censoring.

Parameter* Erasmus MC
Warwick
Model E0
(95% CI**).
Model E1
(95% CI**)
Model W
1. Fraction of late asymptomatic individuals who develop VL (%) 1.42 (1.00–1.84) 1.42 (1.00–1.84) N/A
2. Duration late asymptomatic stage (days) 69 (49–119) 69 (49–118) N/A
3. Duration early recovered stage (days) 237 (196–299) 236 (196–298) N/A
4. Infectivity of early asymptomatic stage 0*** 0.0114**** N/A
5. Infectivity of late asymptomatic stage 0*** 0.0229 (0–0.0533) N/A
6. District specific average sandfly-to-human ratio (SHR) Saharsa 3.92 0.604 0.445
East Champaran 2.20 0.392 0.381
Samastipur 2.49 0.401 0.398
Gopalganj 2.33 0.403 0.390
Begusarai 2.78 0.383 0.425
Khagaria 2.44 0.388 0.401
Patna 2.12 0.359 0.380
West Champaran 1.84 0.351 0.364
7. IRS efficacy*****
(Employing data from all 8 districts)
0.999 0.829 0.006

NB. Values for models E0 and E1 presented here are only for the duration of early asymptomatic stage of 202 days; values for the other sub-models are listed in Table S3 of Supplementary File 3.

*

Parameters 1–5 were fitted to the KalaNet data (models E0 and E1), parameter 6 was fitted to the Thakur data (models E0 and E1) and the CARE data (model W), and parameter 7 was fitted to the CARE data (models E0, E1 and W).

**

Confidence interval.

***

Pre-set values, in model E0 asymptomatic individuals are considered not to be infective.

****

Not fitted, but calculated as half the infectivity of the late asymptomatic stage.

*****

The IRS efficacy is multiplied by the district-specific IRS coverage rate to get the IRS impact on the SHR. Note, however, that the dependence of the SHR on the IRS efficacy is linear in models E0 and E1, but exponential in model W (see Additional File 1 of (Le Rutte et al., 2016) and Supplementary File 1).