Biology Reference
In-Depth Information
Figure 1.2
Regional tiles of
Plasmodium vivax
endemic countries.
The regional tiles
shown were used to stratify the modelling of
P. vivax
endemicity. Pink = the Americas;
blue = Africa+; green = Asia; orange = Asia-Pacific; grey = non endemic for
P. vivax
. For
interpretation of the references to colour in this figure legend, the reader is referred to
the online version of this topic.
3.1.
P. vivax
Malaria Limits and Endemicity
Figure 1.3
shows estimated limits of unstable (light grey) and stable (dark
grey) transmission of
P. vivax
in Panel 1 (A1-D1) and, within Panel 2 (A2-
D2), levels of endemicity (blue to yellow to red, red being >7% prevalence).
The map is accompanied by two measures of uncertainty in our endemic-
ity estimates: one absolute measure and the second weighted by population
density in each location (
Fig. 1.4
).
A detailed review of the data and methodology used to create this map
is given at the end of this chapter in Section 6. In brief, we have used 9970
spatiotemporally unique records of parasite prevalence (
Fig. 1.3, Panel 1
), envi-
ronmental covariates and a Bayesian geostatistical model with seasonal and age-
standardisation components to generate a smooth map of point estimates of
parasite prevalence in all areas with stable transmission (
Gething et al., 2012
).
Throughout the world, endemicity was predicted within a relatively narrow
range, with the point estimate of the
P. vivax
PR age-standardised to the 1-99
year age range (
Pv
PR
1-99
) rarely exceeding 7%. Here, we highlight important
aspects of the methodological approach taken that improve on preceding global
estimates of
P. vivax
malaria risk, along with key assumptions and constraints:
Temporal influence
. The maps presented (Figure 1.3, Panel 2) are of ende-
micity in 2010 but the data used to inform the model ranged from 1985 to
2010. The model was designed to down-weight older data, so that whilst they
can help inform the contemporary estimates, particularly in areas lacking
contemporary data, these had less influence on our estimates of prevalence.
This was reflected by higher uncertainty those areas reliant on older data.
