Galveston Bay Recovery Study, 2008-2010 (ICPSR 34801)

Published: Jun 21, 2016

Principal Investigator(s):
National Center for Disaster Mental Health Research; Sandro Galea, Columbia University; Fran Norris, United States Department of Veterans Affairs. National Center for PTSD, and National Center for Disaster Mental Health Research

Version V1

The Galveston Bay Recovery Study (GBRS) was designed to study trajectories of wellness after Hurricane Ike hit the Galveston Bay area on September 13, 2008. The sample included adults who were living in Galveston County or Chambers County, Texas at the time of the hurricane, not just those who remained in the area after the hurricane, who may have been less affected by the storm. Three interviews were conducted approximately 2-5, 5-9, and 14-18 months after the hurricane, respectively. Information was obtained on experiences during Hurricane Ike, lifetime traumatic events, and mental health and functioning before and after the hurricane, as well as between survey waves (including assessment of posttraumatic stress disorder, depression, generalized anxiety disorder, panic disorder, and suicidality). Demographic variables include race/ethnicity, age, education, marital status, number of children/offspring, income, and employment status.

National Center for Disaster Mental Health Research, Galea, Sandro, and Norris, Fran. Galveston Bay Recovery Study, 2008-2010. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2016-06-21.

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote

United States Department of Health and Human Services. National Institutes of Health. National Institute of Mental Health (NIMH 5 P60 MH082598)


Public and restricted versions of the data are included in this collection. Numeric variables with sensitive information has been recoded using top and bottom coding methods to protect respondent confidentiality in the public version of the data. Due to the sensitive nature of the restricted data, users will need to complete a Restricted Data Use Agreement before they can obtain the restricted version. These forms can be accessed on the download page associated with this dataset.

2008 -- 2010

2008-11-07 -- 2009-03-24

2009-02-06 -- 2009-06-29

2009-11-19 -- 2010-04-13

File set 8 "Combined Imputation Data (Restricted Use)" is a merged file containing various imputations of the original data with sensitive information. The variables are differentiated by the suffix "_it#" in the variable names (where # represents the imputation number).

ICPSR has generated a numerical sequential case identification number called CASEID "Case Identification Number" because the original variable was alphanumeric. The original CASEID has been renamed CASEID_ORIG "Original Case Identification Number" and can be used for merging purposes.

Geographic information (zip code, census tract, and census block) was removed from the data to preserve confidentiality (note that these variables still appear in the original study codebook).

The study conducted multiple imputation using the Sequential Regression Imputation Method implemented in IVEWARE. Further information about the missing data and multiple imputation is included in Appendix #4 in the study codebook.

There are eight datasets included in this study: Part one contains the public version of the original study data, including missing values. Parts two through six contain imputed values in place of missing data. Part seven contains the restricted version of the original study data and part eight is a combined dataset that is comprised of the restricted variables from parts two through six.

The primary objective of the study was to document post-disaster trajectories of mental health and functioning and to evaluate the influence of socio-demographic characteristics, pre-disaster resources, and disaster experiences on those trajectories.

The study area (Galveston county and Chambers county, Texas) was divided into five strata based on FEMA maps of flooding after Hurricane Ike as well as United States Census data on poverty levels in the year 2000. Differential sampling was employed across strata to ensure the inclusion of residents of the most affected areas and those most likely to have experienced hurricane-related traumatic events. The five strata were as follows, with Stratum 1 expected to have experienced the greatest hurricane damage and Stratum 5 the least damage: (1) Galveston Island and the Bolivar Peninsula, which suffered storm surge damage; (2) Flooded areas of the mainland; (3) Non-flooded areas of the mainland with relatively high rates of poverty (i.e., over 15 percent of the population) according to 2000 United States Census data; (4) Non-flooded, non-poverty areas east of Route 146; and (5) Non-flooded, non-poverty areas west of Route 146 and the rest of Chambers county. A map of the study area indicating the five strata is included in the following publication: Tracy M, Norris F, Galea S. Differences in the determinants of posttraumatic stress disorder and depression after a mass traumatic event. Depression and Anxiety 2011; 28: 666-675 (Figure 1, p. 668). Eighty segments consisting of clusters of Census blocks were selected from the five strata. Addresses of all housing units in those eighty segments prior to Hurricane Ike were obtained from a database maintained by the Experian credit reporting agency, as well as the United States Postal Service (USPS) Delivery Service files for the segments. These address lists were supplemented with field listings by staff from the University of Michigan Survey Research Center, and 2,263 households were selected from the final address listings. Upon contact with the household, a complete enumeration of adult household members was obtained and the respondent was randomly selected from all eligible household members.

Cross-sectional ad-hoc follow-up


The target population for this study was all adults (18 years of age or older) who were living in Galveston county or Chambers county on September 13, 2008, when Hurricane Ike made landfall, and who had been living in the area for at least one month prior to the hurricane.


survey data

computer-assisted personal interview (CAPI)

computer-assisted telephone interview (CATI)

658 individuals completed Wave 1 interviews, for a screening/locating rate of 0.52, a cooperation rate of 0.83, an eligibility rate of 0.81, and a response rate (AAPOR RR4) of 0.40. Of those 658 baseline participants, 529 (80 percent) completed a Wave 2 interview and 487 (74 percent) completed a Wave 3 interview.

Several well-established scales were used to assess mental health symptoms and functioning after exposure to Hurricane Ike and between survey waves, including the PTSD Checklist-Civilian version (PCL-C), the Short Posttraumatic Stress Disorder Rating Interview - Expanded version (SPRINT-E), the STRS Checklist, the Peritraumatic Dissociative Experiences Questionnaire (PDEQ), the Patient Health Questionnaire-9 (PHQ-9), and the Mini-International Neuropsychiatric Interview (MINI). A list of citations that include descriptions, coding information, and reliability and validity information for these scales can be found in Appendix #1 of the study codebook.



2016-06-21 ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:

  • Created variable labels and/or value labels.
  • Created online analysis version with question text.
  • Performed recodes and/or calculated derived variables.
  • Checked for undocumented or out-of-range codes.

Weights were calculated to account for differential sampling probabilities across the five study strata, probabilities of selection within households, and nonresponse. An additional post-stratification adjustment was applied to match the sample to the population in Galveston and Chambers counties according to the 2005-2007 American Community Survey (ACS). Finally, samples participating in Waves 2 and 3 were weighted further to account for attrition. More information about the calculation and application of weights in the GBRS is available in Appendix #3 of the study codebook.


  • Data in this collection are available only to users at ICPSR member institutions.

  • One or more files in this data collection have special restrictions. Restricted data files are not available for direct download from the website; click on the Restricted Data button to learn more.

RCMD logo

This study is provided by Resource Center for Minority Data (RCMD).