Assessment of Household Preparedness Through Training Exercises — Two Metropolitan Counties, Tennessee, 2011

Weekly
September 14, 2012 / 61(36);720-722

Public health emergency preparedness involves improving both workforce and household capacity to manage disasters. To improve preparedness at both levels, the Tennessee Department of Health (TDH) formed a Rapid Assessment of Populations Impacted by Disasters (RAPID) team. In 2011, the team used Community Assessment for Public Health Emergency Response (CASPER) two-stage cluster sampling methodology to measure household preparedness for disasters or emergencies in two metropolitan counties. In the two counties, 23% and 31% of households reported being "well-prepared" to handle disasters or emergencies, 43% and 44% reported being "somewhat prepared," and 25% and 20% reported being "not at all prepared." As a result of this experience, RAPID teams were able to improve their methods, streamline processes, and create a better community assessment toolkit. To increase preparedness at both the community and workforce levels, public health departments should assess community preparedness to inform the planning process and provide field training and exercise opportunities for public health workers.
Public health preparedness programs are charged with building capability to respond to disasters and improving community preparedness (1). To help achieve these goals, TDH formed a RAPID team in 2010 made up of local and state health department staff members. Using CASPER methods, the RAPID team, in conjunction with two metropolitan county health departments and volunteers, conducted community assessments in 2011 to 1) assess household-level preparedness in two major metropolitan areas in Tennessee, 2) evaluate the field team deployment process, and 3) train and exercise the emergency response workforce in postdisaster survey methods. Although CASPER methodology has been used and validated during disaster response (2) and nondisaster scenarios (3), assessing community preparedness is a new use for these methods.
For both surveys, RAPID team leaders with previous experience conducting CASPER surveys used materials from a rapid community assessment toolkit developed by TDH. The toolkit was intended to facilitate the field survey deployment process by providing premade training in methodology, safety and logistics, plus maps of sampling areas, logistics forms, census data, databases, and survey templates. The two surveys provided opportunities to test and improve several of these tools.
Knox County, which includes the city of Knoxville, had a 2010 population of 423,237. Davidson County, which includes the city of Nashville, had a 2010 population of 626,681 (4). The two counties were selected for these exercises because their health departments expressed interest in assessing community preparedness. Before deployment to the field to conduct household interviews, approximately 80 public health workers, university students, and volunteers attended 4 hours of training. The trainings included modules on 1) basic concepts of CASPER methodology, 2) team logistics (e.g., communications, navigation in the field), 3) use of survey and household tracking forms, and 4) personal safety procedures while in the field. Before going into the field, teams conducted mock interviews using the survey questionnaire to ensure interview consistency and understanding of the survey aims. Rosters of those completing the training and field portions of the exercises were retained by the health departments to facilitate redeployment of experienced personnel during future rapid community assessments.
Three broad categories of household preparedness were assessed: 1) preparedness planning and supplies, 2) communication and information sources during disasters, and 3) the presence of household pets and pet-related evacuation plans. Survey questions were derived from the Behavioral Risk Factor Surveillance System (5) General Preparedness module. The Davidson County survey included a question asking about a written or spoken preparedness plan, whereas Knox County asked about a written evacuation plan. The CASPER two-stage cluster sampling design was used to obtain representative samples of approximately 210 households for interviews for each assessment. In the first stage, the RAPID team randomly selected 30 census blocks within each county, with probability of selection proportionate to population size (6). In the second stage, households were selected using sequential sampling, starting in the visually estimated center of each identified cluster. At each household that participated, one adult representative was interviewed. Of 252 Knox County and 316 Davidson County homes at which a resident answered the door, 197 (78%) and 184 (58%) interviews, respectively, were completed; residents at the remaining households declined participation.
Survey Results
Results of the interviews indicated that, in Knox County, 23% of respondents reported being "well-prepared" to handle disasters or emergencies, 43% reported being "somewhat prepared," and 25% reported being "not at all prepared" (9% responded: "don't know"). In Davidson County, the corresponding percentages were 31%, 44%, and 20% (5% responded: "don't know") (Table). In Knox County, 11% reported having a written disaster evacuation plan; in Davidson County, 40% reported having either a written or spoken plan for emergencies.
When asked about household supplies, 78% (Knox) and 87% (Davidson) of households reported having a working flashlight with batteries, and 60% (Knox) and 55% (Davidson) reported having a battery-operated radio with batteries. Eighty-four percent (Knox) and 82% (Davidson) of households reported having 3-day supplies of nonperishable food, 39% (Knox) and 54% (Davidson) reported 3-day stores of water, and 74% (Knox) and 91% (Davidson) reported 3-day supplies of prescription medications.
Mobile telephones were reported as the primary means of communicating during disasters in 83% (Knox) and 90% (Davidson) of households (Table). Television was reported to be the primary means of getting information during disasters in 45% (Knox) and 46% (Davidson) of households, with radio the primary source for 23% in both counties. Pet or livestock ownership was reported in 49% (Knox) and 47% (Davidson) of households (Table). During a disaster requiring evacuation, 63% (Knox) and 87% (Davidson) planned to bring their pets, whereas only 3% (Knox) and 4% (Davidson) anticipated leaving their pets or livestock with food.
Assessment Improvements
Based on the experience in these two surveys, the RAPID teams were able to revise and improve training on household selection in the field, navigation in the community using cluster maps, and use of associated tracking forms. The household selection training revisions include visual examples of cluster maps with displayed starting points and discussions pertaining to team strategies for sequential sampling in areas with dispersed households. To streamline future responses, the teams refined cluster map templates with navigation maps, added navigation elements to the premade presentations contained in the toolkit, and refined tracking and sign-in forms to efficiently manage team logistics. The teams strengthened the safety module with information on team identification in the field and routine communications with logistics personnel. The toolkits now contain all this information on single DVDs and have been distributed to public health emergency response teams in all regions of Tennessee. The improved kits have the potential to reduce the response time for community assessments.

Reported by

Kathleen C Brown, PhD, Knox County Health Dept, Knoxville; Nancy Horner, Metro Public Health Dept, Nashville; Melissa Fankhauser, MPH, Tennessee Dept of Health. Joseph Roth, Jr., MPH, Career Epidemiology Field Officer, Office of Science and Public Health Practice, Office of Public Health Preparedness and Response; Tristan Victoroff, MPH, Scientific Education and Professional Development Program Office, Office of Surveillance, Epidemiology, and Laboratory Svcs, CDC. Corresponding contributor: Joseph Roth, Jr., jroth@cdc.gov, 615-253-8669.

Editorial Note

Quantifying household-level preparedness provided planners the information needed to guide targeted program activities. Although public health emergency preparedness programs in each region continually bolster response plans, enhanced efforts to improve household preparedness are needed because substantial numbers of households report being unprepared or less than fully prepared for disasters. Based on the survey results in this report, plans for public health or other emergency messaging during a disaster should include television, radio, and the Internet because those are the primary means of obtaining information during a disaster for more than three fourths of those surveyed. Approximately half the population surveyed owned pets or livestock, and most owners plan to bring their pets during evacuations. Therefore, shelter and evacuation plans need to accommodate both evacuees and their animals, and plans for managing livestock should be made by the appropriate agencies. Planners should incorporate these findings into response strategies and coordinate community messaging both to improve household preparedness and to guide community actions.
During the survey periods, press releases pertaining to the surveys prompted print media and television dissemination of information in each county regarding preparedness and the role of public health, enabling planners to reinforce the CDC Get a Kit, Make a Plan, Be Prepared (7) themed messages to the communities. Moreover, the surveys provided the opportunity for direct interaction between public health staff members and hundreds of members of the community to discuss household preparedness strategies and distribute preparedness guidance.
The findings in this report are subject to at least four limitations. First, interviews were conducted in the daytime and early evenings, so households where the adults were working outside the home at the time of interviews had limited representation. Selection bias might have occurred if households were selected based on occupancy or interviewer safety concerns. Second, results are representative only of the populations in the two counties surveyed, thus limiting the ability to generalize to other regions in Tennessee or elsewhere. Third, all information was self-reported, and respondents were not required to present any evidence that a preparedness measure (e.g., 3-day supply of water or a working flashlight) had been met. Finally, widespread media coverage might have contributed to social desirability bias, resulting in an overestimate of households reporting that they are well-prepared or somewhat prepared.
These surveys provided valuable information about community preparedness in two of Tennessee's largest metropolitan areas, while providing staff members, students, and volunteers with experience conducting the type of surveys that will need to be done in postdisaster settings, when response demands can exceed normal operating capacity. CASPER surveys during emergency responses require organization of multiagency participation (e.g., public health, emergency response, and nongovernmental organizations), application of scientifically sound survey methods, and prompt data collection, analysis, and reporting to inform response activities. Knox County had not conducted such a field exercise previously, and Davidson County had conducted only one previous CASPER survey. By conducting surveys during nondisaster situations, both jurisdictions and team members gained valuable experience that can facilitate survey planning and implementation in future emergency responses. The field experience knowledge and confidence gleaned from successfully completing these exercises will better enable staff members to conduct community assessments during future emergency responses.

References

CDC. Public health preparedness capabilities: national standards for state and local planning, March 2011. Atlanta, GA: US Department of Health and Human Services, CDC; 2011. Available at http://www.cdc.gov/phpr/capabilities/dslr_capabilities_july.pdf . Accessed September 12, 2012.
Malilay J, Flanders WD, Brogan D. A modified cluster-sampling method for post-disaster rapid assessment of needs. Bull World Health Organ 1996;74:399–405.
CDC. Intent to receive influenza A (H1N1) 2009 monovalent and seasonal influenza vaccines—two counties, North Carolina, August 2009. MMWR 2009;58;1401–5.
US Census Bureau. State and county quickfacts. Washington, DC: US Census Bureau; 2012. Available at http://quickfacts.census.gov/qfd/maps/tennessee_map.html , Accessed September 7, 2012.
CDC. Behavioral Risk Factor Surveillance System 2007 questionnaire. Atlanta, GA: US Department of Health and Human Services, CDC; 2006. Available at http://www.cdc.gov/brfss/questionnaires/pdf-ques/2007brfss.pdf . Accessed September 7, 2012.
US Census Bureau. State and county estimates for 2009. Washington, DC: US Census Bureau; 2010. Available at http://www.census.gov/did/www/saipe/data/statecounty/data/2009.html . Accessed September 7, 2012.
CDC. Emergency preparedness and response: emergency preparedness and you. Atlanta, GA: US Department of Health and Human Services, CDC; 2012. Available at http://emergency.cdc.gov/preparedness. Accessed September 7, 2012.