The Effects of the April 2011 Tornado Outbreak on Personal Preparedness in Jefferson County, Alabama Lisa C. McCormick, DrPH; Jesse Pevear III, MSPH; Andrew C. Rucks, PhD; Peter M. Ginter, PhD rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr
Objective: The purpose of this study was to examine the effects of a tornado disaster on the personal preparedness of local residents to determine (1) to what extent the tornado outbreak experience had altered preparedness awareness, willingness to act, and levels of personal preparedness of residents as measured by possession of a preparedness kit; and (2) what effect this experience had on the variables associated with having a complete disaster preparedness kit. Design: Two random digit–dialed surveys were completed following the Behavioral Risk Factor Surveillance System protocols. The pre–tornado survey was conducted between October and December 2010 and the post–tornado survey was conducted between January and March 2012. Results: After the April 2011 tornado outbreak, 86.08% of the respondents (n = 1364) reported that they had thought more about personal or family preparedness and 59.65% (n = 907) reported that they had taken actions to increase their level of preparedness. Overall, general awareness of preparedness media campaigns increased significantly (almost 24%; P < .0001), as did the percentage of those having a complete disaster preparedness kit (a 66% increase, not quite doubled from 2010 to 2012; P < .0001). Conclusions: Findings of the study indicate that the disaster had a significant impact on the local residents’ (1) awareness of preparedness campaigns, (2) awareness of the need to be prepared, (3) willingness to become better prepared, and (4) possession of a disaster and emergency preparedness kit and its associated items. KEY WORDS: disaster, disaster experience, disaster
preparedness kit, public health preparedness, tornado
J Public Health Management Practice, 2014, 20(4), 424–431 C 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins Copyright
From April 25 to April 28, 2011, Alabama, Arkansas, Georgia, Mississippi, and Tennessee experienced 351 tornados. Fifteen of these tornados registered 4 or 5 on the Enhanced Fujita (EF) scale and resulted in 338 reported fatalities, making the outbreak the third deadliest on record.1 Overall, the National Weather Service confirmed 29 tornados with more than 1500 injuries and 85 fatalities in central Alabama alone and 2 tornados, an EF-2 and an EF-4,2 in Jefferson County (Birmingham, Alabama), the most populous county in the state.3 These weather events initiated a Major Disaster Declaration on April 28, 2011.4 The purpose of this study was to examine the effects of the April 2011 tornado outbreak experience on the population’s subsequent disaster awareness and personal/household preparedness. Previous studies reporting the effects of disaster experience on population awareness and the level of individual and family preparedness have been inconsistent. Several studies found that disaster experience is not a factor in increasing preparedness of the population affected by the event.5-7 For example, many coastal property owners affected by Hurricane Sandy, although flooded repeatedly in the past, ignored federal incentives and did little to protect themselves from flood damage.8,9 However,
Author Affiliations: Department of Health Care Organization and Policy (Drs McCormick, Rucks, and Ginter) and Survey Research Unit (Mr Pevear), University of Alabama at Birmingham School of Public Health. The 2010 survey was funded by the American Reinvestment and Recovery Act through a cooperative agreement to the Alabama Department of Public Health administered by the Centers for Disease Control and Prevention (CDC). The 2012 survey was funded by the Alabama Department of Public Health. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the views of the Alabama Department of Public Health or the CDC. No author has declared any financial, consultant, institutional conflicts of interest, nor are they involved in other relationships that will lead to conflicts of interest. Correspondence: Lisa C. McCormick, DrPH, Department of Health Care Organization and Policy, University of Alabama at Birmingham, RPHB 330, 1720 Second Ave S, Birmingham, AL 35294 ([email protected]). DOI: 10.1097/PHH.0b013e3182a45104
424 Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Effects of the April 2011 Tornado Outbreak on Personal Preparedness
other studies have found that past disaster experience is an important factor in influencing an individual’s level of preparedness.10-12 Coulston and Denny12 found that exposure to flooding increased preparedness levels among those affected for up to 18 months after experiencing an event. Similarly, Tekeli-Yes¸il et al, in a 2010 study, found that disaster experience was one of the leading factors associated with an individual taking preparedness actions after an earthquake.10 This study adds to the evidence that disaster exposure experience increases the population’s general awareness of preparedness campaigns, awareness of the need to be prepared, willingness to become better prepared, and the levels of personal preparedness as measured by preparedness kits.
● Methods Alabama’s vulnerability to natural disasters led the Alabama Department of Public Health (ADPH) to launch an emergency preparedness public education program (Get10) in the fall of 2007 to promote personal preparedness. The Get10 program was designed to educate the citizens of Alabama on the need to be prepared and give families strategies to be better prepared.13 Get10 promoted a minimum level of household preparedness as 10 essential items in a disaster preparedness kit—water, nonperishable food, manual can opener, 30-day supply of medications, first aid kit, flashlight, battery-powered radio, clothing, personal care items, and important documents.14 The Get10 campaign has continued from 2007 to 2013. Along with awareness and behavioral measures, possession of a disaster preparedness kit was used as a measure of individual and family preparedness. Willingness to take preparedness actions prior to an event and having a disaster preparedness kit were used as measures of an individual’s personal preparedness. In a 2010 survey of Jefferson County (Alabama) residents (prior to the April 2011 storms), approximately 37% of the respondents reported having a complete disaster preparedness kit for use in an emergency. A complete kit was defined as having 9 or 10 of the Get10 items, as not everyone is on medication and needs a 30-day supply. Marital status was the only variable to significantly predict whether a complete kit was on hand.15 After the April 2011 tornado outbreak, a followup survey was conducted to examine the following: (1) to what extent the tornado outbreak experience had altered preparedness awareness, willingness to act, and levels of personal preparedness; and (2) the effect this experience had on the variables associated with having a complete disaster preparedness kit.
❘ 425
Two random digit–dialed surveys were completed using the Behavioral Risk Factor Surveillance System (BRFSS) protocols.16 The pre–tornado survey was conducted between October 15 and December 31, 2010, as part of a modified BRFSS supplement developed by the Centers for Disease Control and Prevention Communities Putting Prevention to Work program.17 The post– tornado survey was conducted between January 1 and March 31, 2012, and was collected specifically to measure changes in the levels of awareness, attitudes, and individual and family preparedness following the tornado outbreak. Both the pre- and post–tornado surveys included 3 sets of questions: (1) a set derived from the yearly BRFSS survey; (2) a set focusing on individual and family preparedness; and (3) a set concerning personal awareness of preparedness strategies and media campaigns. In addition, the post–tornado survey included questions examining individual and family preparedness attitudes and behaviors after the tornado event. The Survey Research Unit (SRU) in the School of Public Health at the University of Alabama at Birmingham was contracted for the data collection. For the last 17 years, the SRU has collected BRFSS data for the ADPH. Internal review board approval was obtained prior to the conduct of each survey (2010 pre–tornado survey: protocol X100907005; 2012 post–tornado survey: protocol X111114011). For both surveys, the SRU followed the Centers for Disease Control and Prevention BRFSS protocol, which requires that respondents be 18 years or older, live in a noninstitutionalized home with telephone access (landline or cellular), and be willing to complete the Computer-Assisted Telephone Interview survey.16 The data for both surveys were weighted using the Centers for Disease Control and Prevention guidelines to ensure that the sociodemographic characteristics of survey respondents represent Jefferson County, Alabama.16 Data were weighted on sex (male: 47.35%; female: 52.65%), age (18-64 years: 82.85%; 65 years and older: 17.15%), and race (white: 53.03%; black: 42.00%; other: 4.97%), using the 2010 US Census data for Jefferson County. All analyses were completed using SAS (version 9.3) survey procedures (SAS PROC SURVEY) that include the cluster, strata, and weighting factors.
Outcome variables Both surveys included items both to determine familiarity with personal preparedness campaigns and to determine which items of a disaster preparedness kit each respondent had on hand to use in an emergency or disaster as recommended by the ADPH Get10 campaign.14 In addition, survey participants were asked specifically whether they had heard, read, or
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
426 ❘ Journal of Public Health Management and Practice seen any advertising on preparing for a natural disaster or other emergency. Both surveys included questions to determine respondent awareness of the ADPH Get10 program. In the post–tornado survey participants were asked (1) whether they thought more about personal or family preparedness since the tornado outbreak; (2) whether they had taken action to increase their level of personal or family preparedness; and (3) whether they had a written family emergency plan.
Descriptive variables Established BRFSS questions were used to determine the demographics of respondents. These measures included sex, age, race, marital status, whether 1 or more children 18 years or younger currently live in the home, education level, and annual household income. Participants were also asked whether they had been previously diagnosed with diabetes and/or angina or coronary heart disease. To determine whether respondents would require a supply of medications in their disaster preparedness kit, they were asked whether they or any household member is currently taking over-thecounter or prescription medications needed on an ongoing basis.
Data analysis Univariate analysis was used to describe study population. Bivariate analysis (ie, χ 2 test) was used to determine whether there were any differences between disaster preparedness kit items on hand and awareness of media campaigns between the 2010 and 2012 surveys. Logistic regression was used to determine which variables most likely contribute to a respondent having a complete kit in 2010 and 2012.
● Results In Table 1, demographics of respondents are shown. The pre–tornado survey had a sample size of 1603 respondents, and the post–tornado survey had a sample size of 1602. There were some significant difference in descriptive variables between the 2010 and 2012 surveys as shown in Table 1, but the data in each survey are weighted to represent the population distribution of Jefferson County, Alabama. In both surveys, the majority of respondents were female, between the ages of 18 and 64 years, non-Hispanic white, no children aged 18 years or younger at home, with at least a high school education, and an annual household income of $25 000 or greater. As shown in Table 2, 86.08% of the respondents (n = 1364) reported that they had thought more about
personal or family preparedness after the April 2011 tornado outbreak. Almost 60% (59.65%; n = 907) reported that they had taken actions to increase their level of personal or family preparedness. However, only 20.71% (n = 375) reported having a written family emergency plan. Table 3 reports that 65.75% of respondents (n = 1004) stated that they had heard, read, or seen advertising on preparing their family for a natural disaster or other emergency in 2012 whereas 41.82% (608) reported awareness in 2010. A χ 2 test found the difference between 2010 and 2012 respondents’ advertising awareness to be significant (P < .0001). In addition, Table 3 shows that awareness of the ADPH Get10 campaign decreased significantly from 2010 to 2012 (P < .0001). The proportion of the population with a complete disaster preparedness kit as well as the respondents who had each kit item suggested by the ADPH Get10 campaign is presented in Table 4 for both the pre- and postsurveys. Chi-square tests were performed on each disaster kit item to determine whether the level of personal preparedness had increased since the April 2011 tornado outbreak. The number of respondents with a complete disaster kit almost doubled from 2010 to 2012 (P < .0001), and there were significant increases in the number of respondents having each one of the Get10 kit items (P < .0001 for each item). Of the 1171 respondents reporting to need medication on an ongoing basis, only 73.97% (n = 888) reported having a 30-day supply of medication in their kit. Among those reporting to be diabetic, 93.70% (274) reported needing medications on an ongoing basis but only 76.25% (233) reported a 30-day supply of medications on hand. Among those with a diagnosis of coronary heart disease or angina, 97.91% (145) reported needing medications on an ongoing basis whereas only 69.29% (110) of those reported having a 30-day supply. Logistic regression was used to determine which demographic variables predict whether someone had a complete kit. As shown in Table 5, in 2010, marital status was the only significant variable in determining whether someone had a complete disaster kit (odds ratio [OR] = 1.685; 95% confidence interval [CI], 1.1212.532). However, in 2012, age (OR = 1.896; 95% CI, 1.228-2.926), marital status (OR = 1.670; 95% CI, 1.0702.605), and annual household income (OR = 1.560; 95% CI, 1.025-2.374) were significant. The McFadden R2 value was calculated to demonstrate goodness of fit (R2 = 0.032).
● Discussion The general effectiveness of preparedness media campaigns is inconclusive and has not been well
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Effects of the April 2011 Tornado Outbreak on Personal Preparedness
❘ 427
TABLE 1 ● Demographic Information
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
Description Sex Male Female Not asked or missing Age categories, y 18-64 65-99 Don’t know/not sure/refused Race White (non-Hispanic) Black (non-Hispanic) Other Don’t know/not sure/refused Marital status Married Not married/widowed/divorced Don’t know/not sure/refused ≥1 children aged 18 y or younger present in home Yes No Don’t know/not sure/refused Education At least high school graduate Did not graduate from high school Don’t know/not sure/refused Annual income ≥$25 000
Objective: The purpose of this study was to examine the effects of a tornado disaster on the personal preparedness of local residents to determine (1) to what extent the tornado outbreak experience had altered preparedness awareness, willingness to act, and levels of personal preparedness of residents as measured by possession of a preparedness kit; and (2) what effect this experience had on the variables associated with having a complete disaster preparedness kit. Design: Two random digit–dialed surveys were completed following the Behavioral Risk Factor Surveillance System protocols. The pre–tornado survey was conducted between October and December 2010 and the post–tornado survey was conducted between January and March 2012. Results: After the April 2011 tornado outbreak, 86.08% of the respondents (n = 1364) reported that they had thought more about personal or family preparedness and 59.65% (n = 907) reported that they had taken actions to increase their level of preparedness. Overall, general awareness of preparedness media campaigns increased significantly (almost 24%; P < .0001), as did the percentage of those having a complete disaster preparedness kit (a 66% increase, not quite doubled from 2010 to 2012; P < .0001). Conclusions: Findings of the study indicate that the disaster had a significant impact on the local residents’ (1) awareness of preparedness campaigns, (2) awareness of the need to be prepared, (3) willingness to become better prepared, and (4) possession of a disaster and emergency preparedness kit and its associated items. KEY WORDS: disaster, disaster experience, disaster
preparedness kit, public health preparedness, tornado
J Public Health Management Practice, 2014, 20(4), 424–431 C 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins Copyright
From April 25 to April 28, 2011, Alabama, Arkansas, Georgia, Mississippi, and Tennessee experienced 351 tornados. Fifteen of these tornados registered 4 or 5 on the Enhanced Fujita (EF) scale and resulted in 338 reported fatalities, making the outbreak the third deadliest on record.1 Overall, the National Weather Service confirmed 29 tornados with more than 1500 injuries and 85 fatalities in central Alabama alone and 2 tornados, an EF-2 and an EF-4,2 in Jefferson County (Birmingham, Alabama), the most populous county in the state.3 These weather events initiated a Major Disaster Declaration on April 28, 2011.4 The purpose of this study was to examine the effects of the April 2011 tornado outbreak experience on the population’s subsequent disaster awareness and personal/household preparedness. Previous studies reporting the effects of disaster experience on population awareness and the level of individual and family preparedness have been inconsistent. Several studies found that disaster experience is not a factor in increasing preparedness of the population affected by the event.5-7 For example, many coastal property owners affected by Hurricane Sandy, although flooded repeatedly in the past, ignored federal incentives and did little to protect themselves from flood damage.8,9 However,
Author Affiliations: Department of Health Care Organization and Policy (Drs McCormick, Rucks, and Ginter) and Survey Research Unit (Mr Pevear), University of Alabama at Birmingham School of Public Health. The 2010 survey was funded by the American Reinvestment and Recovery Act through a cooperative agreement to the Alabama Department of Public Health administered by the Centers for Disease Control and Prevention (CDC). The 2012 survey was funded by the Alabama Department of Public Health. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the views of the Alabama Department of Public Health or the CDC. No author has declared any financial, consultant, institutional conflicts of interest, nor are they involved in other relationships that will lead to conflicts of interest. Correspondence: Lisa C. McCormick, DrPH, Department of Health Care Organization and Policy, University of Alabama at Birmingham, RPHB 330, 1720 Second Ave S, Birmingham, AL 35294 ([email protected]). DOI: 10.1097/PHH.0b013e3182a45104
424 Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Effects of the April 2011 Tornado Outbreak on Personal Preparedness
other studies have found that past disaster experience is an important factor in influencing an individual’s level of preparedness.10-12 Coulston and Denny12 found that exposure to flooding increased preparedness levels among those affected for up to 18 months after experiencing an event. Similarly, Tekeli-Yes¸il et al, in a 2010 study, found that disaster experience was one of the leading factors associated with an individual taking preparedness actions after an earthquake.10 This study adds to the evidence that disaster exposure experience increases the population’s general awareness of preparedness campaigns, awareness of the need to be prepared, willingness to become better prepared, and the levels of personal preparedness as measured by preparedness kits.
● Methods Alabama’s vulnerability to natural disasters led the Alabama Department of Public Health (ADPH) to launch an emergency preparedness public education program (Get10) in the fall of 2007 to promote personal preparedness. The Get10 program was designed to educate the citizens of Alabama on the need to be prepared and give families strategies to be better prepared.13 Get10 promoted a minimum level of household preparedness as 10 essential items in a disaster preparedness kit—water, nonperishable food, manual can opener, 30-day supply of medications, first aid kit, flashlight, battery-powered radio, clothing, personal care items, and important documents.14 The Get10 campaign has continued from 2007 to 2013. Along with awareness and behavioral measures, possession of a disaster preparedness kit was used as a measure of individual and family preparedness. Willingness to take preparedness actions prior to an event and having a disaster preparedness kit were used as measures of an individual’s personal preparedness. In a 2010 survey of Jefferson County (Alabama) residents (prior to the April 2011 storms), approximately 37% of the respondents reported having a complete disaster preparedness kit for use in an emergency. A complete kit was defined as having 9 or 10 of the Get10 items, as not everyone is on medication and needs a 30-day supply. Marital status was the only variable to significantly predict whether a complete kit was on hand.15 After the April 2011 tornado outbreak, a followup survey was conducted to examine the following: (1) to what extent the tornado outbreak experience had altered preparedness awareness, willingness to act, and levels of personal preparedness; and (2) the effect this experience had on the variables associated with having a complete disaster preparedness kit.
❘ 425
Two random digit–dialed surveys were completed using the Behavioral Risk Factor Surveillance System (BRFSS) protocols.16 The pre–tornado survey was conducted between October 15 and December 31, 2010, as part of a modified BRFSS supplement developed by the Centers for Disease Control and Prevention Communities Putting Prevention to Work program.17 The post– tornado survey was conducted between January 1 and March 31, 2012, and was collected specifically to measure changes in the levels of awareness, attitudes, and individual and family preparedness following the tornado outbreak. Both the pre- and post–tornado surveys included 3 sets of questions: (1) a set derived from the yearly BRFSS survey; (2) a set focusing on individual and family preparedness; and (3) a set concerning personal awareness of preparedness strategies and media campaigns. In addition, the post–tornado survey included questions examining individual and family preparedness attitudes and behaviors after the tornado event. The Survey Research Unit (SRU) in the School of Public Health at the University of Alabama at Birmingham was contracted for the data collection. For the last 17 years, the SRU has collected BRFSS data for the ADPH. Internal review board approval was obtained prior to the conduct of each survey (2010 pre–tornado survey: protocol X100907005; 2012 post–tornado survey: protocol X111114011). For both surveys, the SRU followed the Centers for Disease Control and Prevention BRFSS protocol, which requires that respondents be 18 years or older, live in a noninstitutionalized home with telephone access (landline or cellular), and be willing to complete the Computer-Assisted Telephone Interview survey.16 The data for both surveys were weighted using the Centers for Disease Control and Prevention guidelines to ensure that the sociodemographic characteristics of survey respondents represent Jefferson County, Alabama.16 Data were weighted on sex (male: 47.35%; female: 52.65%), age (18-64 years: 82.85%; 65 years and older: 17.15%), and race (white: 53.03%; black: 42.00%; other: 4.97%), using the 2010 US Census data for Jefferson County. All analyses were completed using SAS (version 9.3) survey procedures (SAS PROC SURVEY) that include the cluster, strata, and weighting factors.
Outcome variables Both surveys included items both to determine familiarity with personal preparedness campaigns and to determine which items of a disaster preparedness kit each respondent had on hand to use in an emergency or disaster as recommended by the ADPH Get10 campaign.14 In addition, survey participants were asked specifically whether they had heard, read, or
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
426 ❘ Journal of Public Health Management and Practice seen any advertising on preparing for a natural disaster or other emergency. Both surveys included questions to determine respondent awareness of the ADPH Get10 program. In the post–tornado survey participants were asked (1) whether they thought more about personal or family preparedness since the tornado outbreak; (2) whether they had taken action to increase their level of personal or family preparedness; and (3) whether they had a written family emergency plan.
Descriptive variables Established BRFSS questions were used to determine the demographics of respondents. These measures included sex, age, race, marital status, whether 1 or more children 18 years or younger currently live in the home, education level, and annual household income. Participants were also asked whether they had been previously diagnosed with diabetes and/or angina or coronary heart disease. To determine whether respondents would require a supply of medications in their disaster preparedness kit, they were asked whether they or any household member is currently taking over-thecounter or prescription medications needed on an ongoing basis.
Data analysis Univariate analysis was used to describe study population. Bivariate analysis (ie, χ 2 test) was used to determine whether there were any differences between disaster preparedness kit items on hand and awareness of media campaigns between the 2010 and 2012 surveys. Logistic regression was used to determine which variables most likely contribute to a respondent having a complete kit in 2010 and 2012.
● Results In Table 1, demographics of respondents are shown. The pre–tornado survey had a sample size of 1603 respondents, and the post–tornado survey had a sample size of 1602. There were some significant difference in descriptive variables between the 2010 and 2012 surveys as shown in Table 1, but the data in each survey are weighted to represent the population distribution of Jefferson County, Alabama. In both surveys, the majority of respondents were female, between the ages of 18 and 64 years, non-Hispanic white, no children aged 18 years or younger at home, with at least a high school education, and an annual household income of $25 000 or greater. As shown in Table 2, 86.08% of the respondents (n = 1364) reported that they had thought more about
personal or family preparedness after the April 2011 tornado outbreak. Almost 60% (59.65%; n = 907) reported that they had taken actions to increase their level of personal or family preparedness. However, only 20.71% (n = 375) reported having a written family emergency plan. Table 3 reports that 65.75% of respondents (n = 1004) stated that they had heard, read, or seen advertising on preparing their family for a natural disaster or other emergency in 2012 whereas 41.82% (608) reported awareness in 2010. A χ 2 test found the difference between 2010 and 2012 respondents’ advertising awareness to be significant (P < .0001). In addition, Table 3 shows that awareness of the ADPH Get10 campaign decreased significantly from 2010 to 2012 (P < .0001). The proportion of the population with a complete disaster preparedness kit as well as the respondents who had each kit item suggested by the ADPH Get10 campaign is presented in Table 4 for both the pre- and postsurveys. Chi-square tests were performed on each disaster kit item to determine whether the level of personal preparedness had increased since the April 2011 tornado outbreak. The number of respondents with a complete disaster kit almost doubled from 2010 to 2012 (P < .0001), and there were significant increases in the number of respondents having each one of the Get10 kit items (P < .0001 for each item). Of the 1171 respondents reporting to need medication on an ongoing basis, only 73.97% (n = 888) reported having a 30-day supply of medication in their kit. Among those reporting to be diabetic, 93.70% (274) reported needing medications on an ongoing basis but only 76.25% (233) reported a 30-day supply of medications on hand. Among those with a diagnosis of coronary heart disease or angina, 97.91% (145) reported needing medications on an ongoing basis whereas only 69.29% (110) of those reported having a 30-day supply. Logistic regression was used to determine which demographic variables predict whether someone had a complete kit. As shown in Table 5, in 2010, marital status was the only significant variable in determining whether someone had a complete disaster kit (odds ratio [OR] = 1.685; 95% confidence interval [CI], 1.1212.532). However, in 2012, age (OR = 1.896; 95% CI, 1.228-2.926), marital status (OR = 1.670; 95% CI, 1.0702.605), and annual household income (OR = 1.560; 95% CI, 1.025-2.374) were significant. The McFadden R2 value was calculated to demonstrate goodness of fit (R2 = 0.032).
● Discussion The general effectiveness of preparedness media campaigns is inconclusive and has not been well
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Effects of the April 2011 Tornado Outbreak on Personal Preparedness
❘ 427
TABLE 1 ● Demographic Information
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
Description Sex Male Female Not asked or missing Age categories, y 18-64 65-99 Don’t know/not sure/refused Race White (non-Hispanic) Black (non-Hispanic) Other Don’t know/not sure/refused Marital status Married Not married/widowed/divorced Don’t know/not sure/refused ≥1 children aged 18 y or younger present in home Yes No Don’t know/not sure/refused Education At least high school graduate Did not graduate from high school Don’t know/not sure/refused Annual income ≥$25 000
