Enhanced Dengue Surveillance in Pau da Lima

Between January 1, 2009 and December 31, 2010, we conducted enhanced community-based surveillance to detect patients with laboratory evidence of dengue infection among those seeking medical care for AFI at the only public emergency health unit (São Marcos Emergency Center [SMEC]; 38°26'09"W, 12°55'32"S) serving the Pau da Lima slum community in Salvador, Brazil (Fig 1A). The study site for the community-based surveillance was arbitrarily defined to have common boundaries with census tract territories, allowing use of official social and demographic population data to determine if AFI patients who sought medical attention at SMEC lived within the study site. In 2010, we performed a community survey and found that 84% (284 of 337) of the study site residents seek medical assistance for AFI at SMEC.
According to the 2010 national census, the population of Salvador was 2.7 million and 76,352 (3%) people lived in the Pau da Lima study surveillance site [34 ]. The study site was comprised of 98 census tracts (CTs) in an area of 3.7 km² within the Sanitary District of Pau da Lima, a delimited administrative area with a population of 218,706 in 294 CTs [34 ]. The site’s topography is characterized by valleys and hills, with an elevation range of 60 meters (S1 Table). Population density was >215,000 inhabitants per km² for 75% of the study site’s CTs [34 ]. On average, 71.9% of the families living within the study area had a per capita monthly income lower or equal to the Brazilian minimum wage (R$ 510.00; equivalent to US$289.77, in 2010) (S1 Table) [34 ]. Demographic and socioeconomic characteristics of the study site varied among the CTs. In general, CTs located around the study health unit presented higher population densities per household and higher percentages of younger inhabitants, black population, illiteracy, and poverty (S1 Fig). Lack of sanitation was more frequent among CTs located in the northeast region of the study site (S1 Fig).
The Zoonosis Control Center at the Municipal Secretary of Health conducted vector control actions within the study site, according to the national guidelines for dengue control and prevention [35 ]. Vector control activities included community education on vector control measures and bimonthly household visits for entomological surveys and vector control. These actions were routinely performed throughout the study period, except for three months between August 2 and November 3, 2010, when a strike of the dengue control agents interrupted their activities. Although we informed the Pau da Lima Health District about the participants’ laboratory dengue results, we were not able to provide this information in a timely enough fashion to guide the activities of the Zoonosis Control Center agents.