Fig 1.
Seasonal pattern of dengue, rainfall and temperature.
Fig 2.
Proportion of dengue cases from 2012 to 2017 and deaths in 2016 and 2017 by age group.
Fig 3.
Monthly dengue incidence and mortality for States and Regions from 2012 to 2017.
Left Y axis: Incidence per 100,000 population in descending order of overall case count. Right Y axis: Case fatality rate.
Fig 4.
Township level annual dengue incidence per 100,000 population from 2012 to 2017.
Source and terms of use for shapefile: https://geonode.themimu.info/layers/geonode%3Ammr_polbnda_adm3_250k_mimu.
Fig 5.
Top ten endemic townships with annual dengue incidence per 100,000 population from 2012 to 2017.
Fig 6.
Township level annual dengue case fatality rates from 2012–2017.
Source and terms of use for shapefile: https://geonode.themimu.info/layers/geonode%3Ammr_polbnda_adm3_250k_mimu.
Fig 7.
Decomposition of additive time series of dengue cases from 2012–2017.
From top to bottom: reported incidence, trend, seasonality and random component.
Fig 8.
Reported dengue serotypes from 1999 to June 2018.
Fig 9.
Correlation coefficient of dengue incidence per 100,000 population with various weather variables at different lags for Kachin, Kayah State, Rakhine State, Ayeyarwady Region, Mandalay Region, Sagaing Region, Tanintharyi Region and Yangon Region.
Fig 10.
Lag response curves of overall relative risks of each meteorological variable with monthly dengue incidence per 100,000 population in Yangon Region.
Fig 11.
The maximum RR of dengue incidence and mean temperature adjusting for relative humidity and rainfall in Yangon Region.
A. Association with temperature at 26°C, centered at 28°C. B. Maximum associated risks at lag 3 months across different temperatures.
Fig 12.
National level dengue forecasts using cases as predictor.
Fig 13.
Comparisons of dengue forecasts between incidence and incidence plus weather variables showing the best performing combination of weather variables for each State/Region.
Lines for observed incidence, predicted incidence and 95% confidence interval for predicted incidence are shown.