Page 241 - Contributed Paper Session (CPS) - Volume 3

CPS2003 Bruno de S. et al.

Figure 3: Proportions of risk factors per sex in PTB cases, 2000-2010. In
brackets are presented the absolute number of cases.

3.2 Spatial and temporal analysis
Figure 4 shows a clear spatial pattern, with the Metropolitan Area of
Porto/Upper North (Region I - MAP), Metropolitan Area of Lisbon (Region II -
MAL) and Algarve/Lower Alentejo (Region III) areas (red/darker and black
areas in Figure 4) being the higher risk regions that significantly contribute to
an increase of the PTB incidence rates. On the contrary, it shows some regions
with lower risk in the interior north, center, and Alentejo (lighter areas in Figure
4), that are significantly decreasing PTB incidence rates. The model did not
show any significant unstructured (local) spatial effects (not shown here).

Figure 4: For the period 2000 -2010, (a) Spatial distribution of the posterior
means of the global spatial effect; (b) 95% posterior probabilities. Black areas
on (b) denote municipalities with strictly positive credible intervals; white
areas representing municipalities with strictly negative credible intervals; and

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