229106 A Cluster Analysis on Barriers to Evacuate

Tuesday, November 9, 2010

Yoon Soo Park, MS , National Center for Disaster Preparedness, Mailman School of Public Health, Columbia University, New York, NY
Tasha Stehling-Ariza, MPH , National Center for Disaster Preparedness, Columbia University, New York, NY
Jonathan J. Sury, MPH, CPH , National Center for Disaster Preparedness, Columbia University, New York, NY
David M. Abramson, PhD MPH , National Center for Disaster Preparedness, Columbia University, New York, NY
Background/Purpose: How do individuals perceive barriers that prevent them from leaving immediately when ordered to evacuate? In this study, a section from a national random-digit dial survey of the American public's opinions and behaviors on preparedness was used to cluster barriers that hinder an individual from evacuating in an emergency. The study aimed the following: (1) how do individuals perceive and cluster these barriers and (2) how do they differ when controlling for region, political ideology, and socioeconomic status.

Methods: Clustering and similarity trees are useful techniques used to group related objects and can be implemented to assess how individuals think of psychological concepts. Ten barriers—children, elderly relative or friend, sick relative or friend, pets, belongings, family or friends, inability to leave without help, lack of confidence in the evacuation order, lack of transportation, and security of their home—were asked and fit using the Extended Similarities Trees clustering method.

Results/Outcomes: Individuals perceived the elderly and the sick together as well as children and family as one entity. Furthermore, transportation and inability to leave without help were grouped, while belongings and security of home were also considered together. However, pets and lack of confidence in the evacuation order were left ungrouped with any other barrier. There were variations by region and socioeconomic factors.

Conclusion: The findings of this study provide insight for policymakers on which barriers to expect in different groups of individuals so that more effective interventions can be designed to enhance planning efforts in disaster situations.

Learning Objectives:
- Compare barriers that prevent individuals from evacuating in a disaster when controlling for region, political ideology, and socioeconomic factors. - Examine methods to cluster outcome variables via K-means, Hierarchical Agglomerative Cluster Analysis, and Extended Similarities Trees - Identify overlapping and hierarchical similarities among clusters

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am a Data Manager/Analyst at the National Center for Disaster Preparedness that collected and analyzed the data used for this study. I am also a doctoral candidate in Measurement and Statistics at Columbia University.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.