Disaster epidemiology is a relatively new field that seeks to examine health outcomes in the midst of disasters. Disaster epidemiology was essentially defined in the late 50’s in a review by Harvard professors, Saylor and Gordon, as they sought to apply general epidemiologic procedures to the investigation of disasters both natural and manmade (Armenian, 4). It aims to assess what services are necessary to those immediately affected by disasters and also to study short and long-term factors that influence health during disasters as an aid to policy makers who prepare for future calamities. Two difficulties with disaster epidemiology cited in a review on the subject deal with “characterization of the population at risk and exposed to the disaster” and “estimation of the exposure to the disaster” (Dominici, 9). Two goals of disaster epidemiology cited in another review describe the necessity of “needs assessment” and “improvements in contingency planning for future disasters.” (Van der Berg, 56) In this blog, I will briefly evaluate these two limitations and two goals in relation to a pair of similar articles written on mental health outcomes following Hurricane Katrina.
The first article is titled “Mental Health Service Use among Hurricane Katrina survivors in the Eight Months after the Disaster” by Dr. Wang et al. The article examines mental health service use in Katrina survivors, finding that few of the study subjects received adequate care. Most respondents cited enabling factors (preventative barriers) as the main reason why their care was limited. Disaster epidemiology has difficulty characterizing the population exposed and at risk to the disaster along with estimating the population’s degree of exposure to the disaster. This problem was addressed in the article by using data from a sample of 1043 pre-hurricane residents that were part of a Hurricane Katrina Community Advisory Group (CAG) defined by FEMA as residents directly affected by Katrina living in certain areas in New Orleans and adjacent states. The residents sampled were surveyed through random digit dialing. An initial issue with the baseline surveys is calculating the proportion of respondents to the survey. In this study, an estimated 42% of those contacted responded to the survey, but verification of the true proportion of respondents is impossible given that there is no way to determine which phone numbers were working numbers. This highlights one issue with disaster epidemiology, in that locating and tracking members of the target population can be difficult given the damage to infrastructure and population dispersal that is a natural consequence of disasters. (Wang 1404) Another problem with defining the target population was that telephone calls were used to survey the population. By the author’s own admission, one important limitation of this technique is that “the most disadvantaged and possibly most severely ill people” (Wang, 1410) are underrepresented because they may not have access to telephones. This point underscores another common problem in disaster epidemiology, locating or characterizing those heavily exposed to the disaster. Mental health severity was examined using the validated K6 screening method and health service use was evaluated using reasonably standard methods. Outcome assessment will not be examined in depth as it does not relate specifically to challenges in disaster epidemiology.
The paper found that a significant portion of respondents had abandoned seeking mental health services following the hurricane. Most respondents claimed a lack of enabling factors prevented them from seeking care. Additionally, the authors write that “only 15% of respondents who were directly affected and 36% of respondents with probably posttraumatic stress disorder or depression sought mental health care by six months (Wang, 1408). This finding aids greatly with both needs assessment and contingency planning for future disasters. Policy makers can use this information to strengthen mental health services in the wake of a similar disaster in the future. The authors also found that the majority of Katrina survivors who did seek out mental health services did so through the general medical sector rather than engaging specialty services first. They go on to advise policy makers that specialty health personnel should be integrated into the general medical sector in the event of a disaster so that the disadvantaged can get the care they need. This recommendation demonstrates the importance of disaster epidemiology in contingency planning. The discussion portion of the paper spends a great deal of time giving recommendations to policy makers on how to approach mental health service during disasters in the future.
The second article, by Dr. Kessler et. Al, is titled “Trends in mental illness and suicidality after Hurricane Katrina.” The authors did a cohort study of hurricane residents, surveying their mental health and then tracking developments in their health over two years. A similar method, as in the previous study, was used to characterize the populations. Respondents were selected through random-digit dialing and selected from the 1.4 million families that applied for assistance from the American Red Cross, and also from those families that were housed in FEMA hotels. While the same issues that the earlier study encountered undoubtedly exist in study (calculating the true proportion of respondents and surveying those without access to telephones), I will focus specifically on characterization of exposure in this paper – another difficulty in disaster epidemiology. The authors spend a significant amount of time discussing stresses and mental health issues in the New Orleans Metro Area as compared to surrounding states. One difficulty in making this comparison, however, is that an equal amount of exposure (the amount residents were affected by the hurricane) is assumed in all groups. No true comparison could be made between residents in the metro area and residents in the surrounding states if their exposures were on different strata. Since there is no truly objective way of defining the exposure, all residents who were characterized by the government as affected by the hurricane and who applied for federal support are treated as equally exposed in the study. This issue is common to disaster epidemiology, in that quantifying exposure to a disaster can be problematic.
In terms of needs assessment and contingency planning, the article makes important points that can be valuable to policy makers. The authors found that there was a significant amount of mental health distress in areas outside New Orleans and that policy makers should plan to spread additional services in areas outside the most affected zone in a future disaster. The authors also found that mental health issues were very weakly related to socio-demographic variables, meaning that mental health is a pervasive issue affecting all age groups and demographics in the aftermath of a disaster (Kessler, 10). Finally, by performing a cohort study and evaluating mental health issues 2 years after the hurricane, the authors were able to determine that the prevalence of distress does not drop significantly immediately after a disaster and that assistance should be provided to those affected by a disaster in the long-term.
Works Cited
Armenian, Haroutune. "Epidemiology in War and Disasters."
Dominici, Francesca, Et Al. "Methodological Challenges and Contributions in Disaster Epidemiology." Epidemiologic Reviews (2005): 9-12.
Kessler, Ronald C., Et Al. "Trends in mental illness and suicidality after Hurricane Katrina." Molecular Psychiatry 4 (2008): 374-84.
Van der Berg, Bellis, Et Al. "The Public Health Dimension of Disasters - Health Outcome assessment of Disasters." Prehospital and Disaster Medicine (2008).
Wang, Philip, Et Al. "Mental Health Service Use Among Hurricane Katrina Survivors in the Eight Months after the Disaster." Psychiatric Services 58 (2007): 1403-410.
No comments:
Post a Comment