Page 257 - Special Topic Session (STS) - Volume 2

STS489 Danielle J.R. et al.

Figure 3: Estimated posterior means of the structured spatial effect (left) and the
unstructured spatial effect (right) on the log-odds of anaemia (criss-cross pattern
indicates water bodies; diagonal lines indicates districts with no data).

4. Discussion and Conclusion
Anaemia control measures need to account for the spatial heterogeneity
that is evident in these countries, as well as take into consideration the
potential factors and type of factors contributing to the spatial heterogeneity.
Kenya and Malawi districts showed negative (low) structured spatial effects.
On the hand, districts in Uganda and Tanzania displayed a mix of positive and
negative structured and unstructured spatial effects, with the unstructured
spatial effect being more prominent. Accordingly, efforts in assessing the local
district-specific drivers of childhood anaemia within each country should be
focused on.

References
1. Abdo, N., Douglas, S., Batieha, A., Khader, Y., Jaddou, H., et al.
(2018). The prevalence and determinants of anaemia in Jordan. East
Mediterr Health J, In press.
2. Besag, J., York, J., and Mollie, A. (1991). Bayesian image restoration, with
two applications in spatial statistics. Ann Inst Stat Math, 343:1–20.
3. Kandala, N.-B. and Madise, N. (2004). The Spatial Epidemiology of
Childhood Diseases in Malawi and Zambia. African Population Studies,
Supplement B, 191–218.
4. Lang, S. and Brezger, A. (2004). Bayesian P-Splines. J Comput Graphical
Statist, 13:183–212. Mainardi, S. (2012) Modelling spatial heterogeneity
and anisotropy: child anaemia, sanitation and basic infrastructure in sub-
Saharan Africa, Int J Geogr Inf Sci, 26(3):387–411.

246 | I S I W S C 2 0 1 9