Mapping residual transmission for malaria elimination
- National Institutes of Health, United States
- Indiana University School of Public Health, United States
- Clinton Health Access Initiative, United States
- National Malaria Control Program, Swaziland
- University of Texas Southwestern Medical Center, United States
- University of California, San Francisco, United States
- University of California, San Francisco Benioff Children’s Hospital, United States
- University of Notre Dame, United States
- University of Southampton, United Kingdom
- University of Oxford, United Kingdom
- University of Washington, United States
- Sanaria Institute for Global Health and Tropical Medicine, United States
Figures
Figure 1
Consensus network plot of causal links.
Panel A: Swaziland imported and local malaria cases (green squares and orange diamonds, respectively) are plotted spatially. Local case pairs identified as putative orphaned chains are indicated by red diamonds. A solitary local case also identified as an orphan is identified as a red diamond within a circle. Panel B: Swaziland imported (green line) and local (orange line) malaria cases are plotted in time, aggregated by month. Panel C: The final consensus network plot is displayed. Local cases are plotted as diamonds and imported cases as green circles. The color of each link corresponds to the “strength” of the connection as measured by the number of parameter sets where that link was identified as optimal. Imported cases that were not found to be the “most likely” parent of a local case are not displayed.
Figure 2
Vulnerability, receptivity and malariogenic potential.
Panel A: Extrapolated Rc values for Swaziland using a zero-inflated negative binomial regression. Areas in orange to red indicate locations where Rc is greater than unity. The legend doubles as a histogram indicating the number of individuals (on a log10 scale) that live within each range of Rc values. Panel B: Extrapolated importation probabilities for Swaziland using a logistic regression. Panel C: Malariogenic potential for Swaziland calculated as the product of Rc and the probability of importation.
Figure 3
Vulnerability, receptivity and malariogenic potential (2010-6/2012).
Panel A: Extrapolated Rc values for Swaziland using a zero-inflated negative binomial regression. Areas in orange to red indicate locations where Rc is greater than unity. The legend doubles as a histogram indicating the number of individuals (on a log10 scale) that live within each range of Rc values. Panel B: Extrapolated importation probabilities for Swaziland using a logistic regression. Panel C: Malariogenic potential for Swaziland calculated as the product of Rc and the probability of importation.
Figure 4
Vulnerability, receptivity and malariogenic potential (7/2012-2014).
Panel A: Extrapolated Rc values for Swaziland using a zero-inflated negative binomial regression. Areas in orange to red indicate locations where Rc is greater than unity. The legend doubles as a histogram indicating the number of individuals (on a log10 scale) that live within each range of Rc values. Panel B: Extrapolated importation probabilities for Swaziland using a logistic regression. Panel C: Malariogenic potential for Swaziland calculated as the product of Rc and the probability of importation.
Figure 5
Timing of ‘orphan’ cases.
The average number of cases per month and total occurrence of looped (or ‘orphaned’ cases) are plotted against month.
Figure 6
Spatial covariates for malaria receptivity regression.
The four significant covariates for the malaria receptivity regression were (A) distance from paved roads, (B) distance from unpaved roads, (C) distance from feeder roads, and (D) distance from Mozambique. All distances were in meters.
Figure 7
Spatial covariates for malaria importation regression.
The ten significant covariates for the malaria importation regression were (A) elevation, (B) population, (C) annual mean temperature (bio1 - http://www.worldclim.org/bioclim), (D) maximum temperature of the warmest month (bio5 - http://www.worldclim.org/bioclim), (E) minimum temperature of coldest month (bio6 - http://www.worldclim.org/bioclim), (F) precipitation of the wettest month (bio13 - http://www.worldclim.org/bioclim), (G) precipitation of driest month (bio14 - http://www.worldclim.org/bioclim), (H) TWI, (I) normalized difference vegetation index, and (J) enhanced vegetation index.
Tables
Table 1
Zero-inflated negative binomial regression summary.
| Factor (source) | Count model coefficient | Zero-inflated coefficient |
|---|---|---|
| Intercept | 7.686407 | 1642.7199 |
| Elevation (m) (http://www.worldclim.org/bioclim) | −0.0026 | −0.65197 |
| Population (http://www.worldpop.org.uk/) | −0.017571 | −0.01190 |
| Annual Mean Temperature (0.1°C) (http://www.worldclim.org/bioclim) | 0.141979 | 30.70232 |
| Max Temperature of Warmest Month (0.1°C) (http://www.worldclim.org/bioclim) | −0.113297 | −23.66837 |
| Min Temperature of Coldest Month (0.1°C) | −0.029091 | −10.04161 |
| Precipitation of Wettest Month (mm) (http://www.worldclim.org/bioclim) | 0.008032 | 0.50592 |
| Precipitation of Driest Month (mm) (http://www.worldclim.org/bioclim) | −0.108767 | 12.04175 |
| TWI | −0.024820 | −4.02392 |
| NDVI (https://landsat.usgs.gov/) | 2.461314 | −159.39390 |
| EVI (https://landsat.usgs.gov/) | −3.732795 | 82.72595 |
| Log(theta) | −0.613861 | NA |
Table 2
GAM logistic regression summary.
| Factor | edf | Chi.sq | p-value |
|---|---|---|---|
| Population (http://www.worldpop.org.uk/) | 6.729 | 688.01 | <2e-16 |
| Paved roads (source: country) | 5.909 | 172.49 | <2e-16 |
| Unpaved roads | 1.002 | 15.88 | 6.88e-5 |
| Feeders roads | 6.499 | 50.37 | 3e-8 |
| Distance to Mozambique (http://www.fao.org/geonetwork/srv/en/main.home) | 7.516 | 75.27 | 1.04e-12 |
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Mapping residual transmission for malaria elimination
eLife 4:e09520.
https://doi.org/10.7554/eLife.09520
