Outdoor Air Pollution in Dakar city (Senegal) and Health Risk Assessment: A pilot study

Abstract Outdoor air pollution is the underlying cause of 4.2 million premature deaths per year, of which some 18% were due to chronic obstructive pulmonary disease (COPD) and 23% due to acute lower respiratory tract infections (IRI). In West African countries, anthropogenic sources of air pollution are concentrated near residential areas, thus exposing populations to high risks of adverse effects. In order to stimulate the implementation of policies to reduce population exposure, this study aimed to describe outdoor air pollution in Dakar (Senegal) and to estimate its health impact. The description was made by comparing the annual averages of various pollutants (NO 2 , SO 2 , PM 10 and PM 2.5 ) to guideline values ​​and on the other hand by converting the monthly averages into time-series which trends and seasonality are described using the R software. The impact was estimated with AirQ + software by calculating the number of specific health outcomes (hospital admissions for respiratory diseases and all-cause mortality) from short-term exposure to outdoor air pollutants’ levels above the threshold of 10 ug/m 3 . Monthly levels follow a linear trend (except that of NO 2 ), with a seasonal component. Pollution peaks are always observed in the December-January-February quarter, while the lowest levels are observed in the July-August-September quarter. Moreover, the annual levels are all above the guideline values, except for SO 2 . Between 2016 and 2018, and for all ages combined, 1379 (13.38%) cases of hospital admission for respiratory diseases can be attributed to short-term exposure to PM 10 . The latter is also the underlying cause of 315 (7.48%) non-accident deaths noted in 2017. Air pollution is a real problem in Dakar in view of the level of indicators. With regard to their health impact, a more exhaustive and more robust estimation would be very useful for better exposure reduction policies. To this end, air quality monitoring and health data management could be improved.