Page 242 - Contributed Paper Session (CPS) - Volume 3

CPS2003 Bruno de S. et al.
grey areas represent municipalities of non-significant effects for PTB
incidence rates (credible intervals containing zero)
Regarding time, Figure 5 shows a slightly non-linear decreasing effect
between 2000 and 2010, confirming the capacity of the model to pick up the
decreasing effect of PTB new cases shown in the previous descriptive analysis
(Figure 1a).

Figure 5: Estimated nonlinear effect of year in PTB incidence rates, together
with 95% credible intervals.

4. Conclusion
Nunes et al. [3] identified two main regions, MAL and MAP, as being high
risk areas for contracting PTB in Portugal in 2001. The results of our study also
suggest a clear urban problem, with MAL (Region II) and the MAP (Region I)
being two of the main areas identified as statistically significant areas of
increased PTB incidence rates (Figure 4). Although with smaller numbers of
new cases of PTB, Algarve and Lower Alentejo (Region III) also emerge as a
region within this category. The metropolitan areas of Lisbon (Region II) and
Oporto (Region I) correspond to two regions with high population density,
resulting immediately in an agglomeration of the main risk groups associated
with high incidence of tuberculosis (e.g. homeless, unemployed, IV drug
addicts and other drugs). On the other hand, Region III which includes Algarve,
not corresponding to an area of high population density throughout the year,
it is associated with seasonal tourism and workers particularly through the
months of April to September, when it also becomes a high density populated
region. It is worth noticing that, after Lisbon with 52% of the total of foreigners
living in Portugal, Algarve, North and Center of Portugal, are the three regions
with the highest percentage of foreigners (13%, each). In addition, 12% of

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