Data Collection Methods during the 2013 Dengue Seroprevalence Survey in Martin County: Lessons Learned from the Field

BACKGROUND: In August 2013, the Florida Department of Health (FDOH) was notified of three locally acquired dengue fever cases in Martin County. By mid-September, 15 more cases had been identified. From September 13 to September 22, FDOH personnel from the state health office and county health departments in Martin and surrounding counties were deployed to conduct a seroprevalence survey of dengue fever in the community. The survey involved three different data collection instruments (household, individual, and business surveys) and two different sampling strategies (random and cluster). This presentation will discuss the challenges and lessons learned from the perspective of the data management team.  

METHODS: To decrease the need for data entry personnel, electronic versions of the questionnaires were provided to the field teams on laptops, and paper versions were provided in the event of equipment failure. Each field team was trained in the use of both the paper version and the electronic version of the questionnaires. Microsoft Access was utilized to build the database for questionnaire responses as well as to create electronic data entry forms to be used in the field. This software was selected due to staff familiarity with the product and unfamiliarity with the data management team’s software of choice, Epi Info 7.  

RESULTS: The data management team experienced several challenges in the use of the electronic form, including a software compatibility issue, rainy weather and non-rugged laptops, time constraints, and high field team turnover rates. Methods implemented to prevent data loss in the field added to the amount of time spent each day aggregating data. Feedback from field teams regarding use of the electronic form was generally neutral with many stating that its use did add time to the interview but saved time and prevented errors associated with delaying data entry to a later time. This time savings allowed data quality checks to be performed earlier in the data management process.  

CONCLUSIONS: An important consideration for future seroprevalence surveys is how to more effectively utilize electronic data collection methods to fully capitalize on their advantages over paper. For example, Epi Info allows data transfer while in the field to update a centralized database, and its use would have mitigated some challenges faced by the data management team. Instead of relying on just-in-time training, it is essential for personnel to be trained in the desired software before an event occurs.