BACKGROUND: Assessing jurisdictional health needs is an integral component to public health and preparedness function. Unfortuantly they are often timely and costly commitments to undertake. The CDC's Community Assessment for Public Health Emergency Response (CASPER) requires minimal financial investment and can be completed in a short timeframe when compared to traditional methods.
METHODS: A standard two-tiered random and weighted sampling was used. Initially, 30 census blocks(clusters) were selected (weighted by population) using a CDC GIS model. Traditionally, after selection, teams are deployed to the clusters to randomly and systematically select 7 households to be surveyed. An inovative, time and cost saving approach was used instead. Using GIS housing data and satellite imagery, 210 households were selected using raster and vector information. Postcards announcing residents of their household being selected as survey participants were distributed prior to the survey event. A combined one day training and survey collection event was held. All survey collections were attempted by teams of two or more. Emergency kits were provided to participating households. If residents were out during collection attempts, cards were left to call and complete survey via telephone.
RESULTS: The method of pre-selecting households saw a cost savings of almost 95% in resources from communication postcards and postage. Without the need of training for housing selection, extended training time was avoided and a time savings of up to 3 hours was found. This savings allowed the training and survey collection event to take place on the same day. 54 motivated volunteers participated on a Saturday morning for training and survey collection. 168 households in 24 clusters were contacted and less than 30 percent initial survey completion rate was achieved. Following the initial event an additional 4 surveys were completed via telephone, and 10 more from weekday evening contact attempts. Survey collection was closed due to season changing and resources.
CONCLUSIONS: When completing a preparedness CASPER, dramatic time and cost savings can be seen by in the effective use GIS to pre-select survey participants. This method would not be effective for completing a CASPER after a disaster event. Considerations for the low primary response rate include: residents were out preparing for or traveling to college football games; residents were out enjoying the 70 degree(F) weather; Oklahoman's are private in nature and don't want to share information. Secondary and tertiary survey attempts would have likely resulted in higher overall survey completion rate.