DISASTERS

There are 2 main types of disasters: natural and technological. Natural disasters occur outside the control of humans, whereas technological disasters are breakdowns in human-made systems.Reference Phifer539Reference Quarantelli40 Immediate threats to public health during the response to a disaster are quite evident. When disaster strikes, the health care system is immediately overwhelmed with injuries and acute illness needs during the initial surge, defined as the sudden increase in demand for emergency medical services. Often, acute illnesses occurring in the initial surge evolve into chronic health needs during the recovery phase of a disaster.Reference Mokdad, Mensah and Posner4 For example, studies have documented long-term medical needs of victims of the World Trade Center collapse due to their extremely high caustic dust exposure,Reference Balmes41Reference Banauch, Hall and Weiden42Reference de la Hoz, Shohet, Bienenfeld, Afilaka, Levin and Herbert43Reference Herbert, Moline and Skloot44Reference Izbicki, Chavko and Banauch45Reference Lin, Jones, Reibman, Bowers, Fitzgerald and Hwang46Reference Mann, Sha, Kline, Breuer and Miller47Reference Mendelson, Roggeveen, Levin, Herbert and de la Hoz48Reference Miller49Reference Moline, Herbert and Nguyen50Reference Oppenheimer, Goldring and Herberg51Reference Prezant52Reference Prezant, Levin, Kelly and Aldrich53Reference Samet, Geyh and Utell54Reference Szeinuk, Padilla and de la Hoz55Reference Wheeler, McKelvey and Thorpe56 of victims of forest fires who had smoke inhalation,Reference Liu, Tager, Balmes and Harrison57Reference Stefanidou, Athanaselis and Spiliopoulou58 and of survivors of the 2004 tsunami who aspirated large quantities of seawater.Reference Bridgewater, Aspinall and Booth59Reference Chierakul, Winothai and Wattanawaitunechai60Reference Kongsaengdao, Bunnag and Siriwiwattnakul61

Disasters pose greater threats to communities lacking resources and access to health care,Reference Mack, Brantley and Bell3Reference Mokdad, Mensah and Posner4Reference Kleinpeter11Reference Ridenour, Cummings, Sinclair and Bixler14Reference Messias and Lacy36Reference Quarantelli40Reference Chen, Keith and Leong62 which are precursors to health care disparities. However, health care disparities are not usually explicitly accounted for in surge capacity modeling.Reference Phillips63 A recent report, “Altered Standards of Care in Mass Casualty Events: Bioterrorism and Other Public Health Emergencies” published by the Agency for Healthcare Research and Quality64 recommended that strategic steps be taken to address community-level factors in response planning, but made no mention of persistent health or healthcare disparities often inherent in these communities.Reference Phillips63

Health and health care disparities are seldom addressed in disaster response and recovery planning. While disaster epidemiology has been recognized as an emerging field with the expansion of human populations into disaster-susceptible regions and global climate change looming on the horizon,Reference Shultz, Russell and Espinel6Reference Ahern, Kovats, Wilkinson, Few and Matthies65Reference Dominici, Levy and Louis66Reference Ebi and Schmier67Reference Noji68Reference Ramirez and Peek-Asa69 the impact of health and health care disparities on disaster epidemiology has not.

There is a paucity of information on the combined effects of a disaster and living in a rural area or other area with existing health or health care disparities on a community's health, access to health resources, and quality of life during the disaster recovery phase. To our knowledge, there has been no report of the burden of extant and incident chronic disease during the disaster recovery phase for a vulnerable community plagued with health disparities in the aftermath of a disaster. Yet, we know that these disparities exist in populations affected by disaster.Reference Krol, Redlener, Shapiro and Wajnberb2Reference Mack, Brantley and Bell3Reference Ford, Mokdad and Link12Reference Ridenour, Cummings, Sinclair and Bixler14 Alternatively, this gap conflicts with the extensive collection of articles published in the lay press during the last 30 years consistently documenting the unmet needs of underserved populations affected by a disaster. The purpose of this article is to document the current status of the science regarding the impact of disaster on health and health care disparities. To accomplish this, we conducted a systematic literature review.

METHODS

We selected the following search terms for our systematic literature review: disaster, health disparities, health care disparities, medically underserved, and rural. By using the combination of these search terms, we performed individual inquiries on widely accepted public health electronic search indices, including the Cambridge Scientific Abstracts, CINAHL Plus, MEDLINE (EBSCO), PubMed-MEDLINE, Web of Science (ISI Citation), Annual Reviews, Applied Social Sciences Index and Abstracts, The Cochrane Library, Health Reference Center-Academic, and Ovid Medline for all relevant articles published in the Unites States from January 1, 1970, to May 15, 2009.

There is considerable uncertainty in the literature regarding exact definitions for health and health care disparities.1570Reference Smedley, Stith and Nelson71 The Health Resources and Services Administration's Index of Medical Underservice has provided a quantifiable way to identify a given area, facility, or population as medically underserved.16Reference Ricketts, Goldsmith, Randolph, Lee, Taylor and Ostermann17 Therefore, we used medical underservice as a proxy for health and health care disparities because we believe that underservice is the leading culprit behind health and health care disparities and better contextualizes the current problem. Therefore, we limited our search to studies conducted in the United States. Our initial inclusion criteria were peer-reviewed, US epidemiological studies published in English that discussed the delayed or persistent effects of health and health care disparities in the midst of a disaster for medically underserved areas. Consequently, owing to the absence of epidemiological studies, we expanded our search to all relevant health studies, including the social sciences and health services research.

To further document the disconnect between health and health care disparities and disaster-related health research, we conducted a minisearch of the lay press using the Lexis-Nexis Academic search database. We used the power search option, narrowing the search within news wire services published from January 1, 1970, to May 15, 2009. Search term categories were as follows: health care and disasters; health care, disasters, and recovery; health care, disasters, and chronic diseases; health care, disasters, and long-term effects; health care, disasters, and rural; health disparities and disasters; and disasters and medically underserved.

RESULTS

There has been extensive research published on disasters, health disparities, health care disparities, and medically underserved populations individually, but few studies addressed these topics collectively (Table 1). We reviewed 12 articles that met the inclusion criteria; however, only 7 pertained to disasters in medically underserved populations (Table 2).Reference Krol, Redlener, Shapiro and Wajnberb2Reference Mack, Brantley and Bell3Reference Mokdad, Mensah and Posner4Reference Hsu, Mas, Jacobson, Harris, Hunt and Nkhoma9Reference Ford, Mokdad and Link12Reference Ridenour, Cummings, Sinclair and Bixler14Reference Fowkes, Blossom, Anderson and Sandrock72 Of those 7, 5 concerned a natural disaster (ie, Hurricane Katrina),Reference Krol, Redlener, Shapiro and Wajnberb2Reference Mack, Brantley and Bell3Reference Mokdad, Mensah and Posner4Reference Ford, Mokdad and Link12Reference Ridenour, Cummings, Sinclair and Bixler14 1 a bioterrorism disaster,Reference Hsu, Mas, Jacobson, Harris, Hunt and Nkhoma9 and 1 concerned natural and technological disasters.Reference Fowkes, Blossom, Anderson and Sandrock72 There were 3 articles reporting studies of medically underserved populations,Reference Krol, Redlener, Shapiro and Wajnberb2Reference Mack, Brantley and Bell3Reference Fowkes, Blossom, Anderson and Sandrock72 and only 1 study explicitly targeted rural medically underserved communities.Reference Hsu, Mas, Jacobson, Harris, Hunt and Nkhoma9 Of the studies, 4 were cross-sectional,Reference Krol, Redlener, Shapiro and Wajnberb2Reference Hsu, Mas, Jacobson, Harris, Hunt and Nkhoma9Reference Ford, Mokdad and Link12Reference Ridenour, Cummings, Sinclair and Bixler14 and the remaining 3 were descriptive.Reference Mack, Brantley and Bell3Reference Mokdad, Mensah and Posner4Reference Fowkes, Blossom, Anderson and Sandrock72 There were no relevant follow-up articles addressing the effects of a disaster on a community with preexisting medical underservice issues or subsequent amplified health care disparities after the disaster. The following is a brief critique of each of the 4 cross-sectional studies.

TABLE 1

Summary of All Search Term Hits by Public Health Index for Literature Review From January 1, 1970, to May 15, 2009

TABLE 2

Key Articles Discussing the Effects of a Disaster on Medically Underserved Populations

Ford et alReference Ford, Mokdad and Link12 conducted a cross-sectional study that explored how an existent surveillance system could be used to estimate chronic disease needs for a natural disaster by using the 2004 Behavioral Risk Factor Surveillance System to estimate prevalence of diabetes, heart disease, stroke, hypertension, and asthma. Cardiopulmonary disease was not included. A quarter of respondents had at least 1 chronic disease, 15.6% had 1 condition, 8.4% had 2 conditions, 1.1% had 3 conditions, and 0.3% had 4 or more conditions. This study revealed that preexisting surveillance systems, such as the Behavioral Risk Factor Surveillance System, can aid disaster response personnel in assessing chronic disease needs among disaster-affected populations. Ford et alReference Ford, Mokdad and Link12 demonstrated that chronic disease management after a disaster is an understudied research priority.

Hsu et alReference Hsu, Mas, Jacobson, Harris, Hunt and Nkhoma9 conducted a cross-sectional study that assessed language, confidence, and training needs in responding to public health emergencies among rural medical providers in Texas. The sampling frame consisted of a physician database supplied by the Texas State Board of Medical Examiners. A semistructured survey was mailed or administered over the Internet to 841 practicing or retired physicians in 37 north Texas counties that assessed language use, perceived confidence in ability to respond to public health emergencies, and training experience. Prior experience in chemical exposure emergencies was reported by 20.9% of respondents. Nearly half were willing to offer their services in a public health emergency. However, 77.5% lacked confidence in their skills to effectively diagnose and treat victims of a public health emergency. This study highlighted the fact that many physicians lack public health emergency awareness, knowledge, and expertise, especially physicians in rural areas.

Krol et alReference Krol, Redlener, Shapiro and Wajnberb2 used an innovative mobile medic approach to conduct a cross-sectional survey that identified the acute and chronic health care needs of medically underserved populations residing in Mississippi after Hurricane Katrina. Data from 2 Children's Health Fund mobile medical units at 23 sites in the Gulfport-Biloxi area included all patient encounters (ie, chief complaint, diagnoses, vaccines dispensed, prescription medications distributed, and referrals) from September 5 through 20, 2005. Out of the 1,187 recorded patient encounters, there were 1,428 documented reasons to visit the mobile medical unit. The top 2 reasons were for vaccinations (n=638 patients with a documented reason to visit) and prescription medication needs (n=149 patients with a documented reason to visit).

Respiratory (27.8%), circulatory (27.8%), and minor injury (19.2%) were the most common diagnoses among all persons surveyed. For people with at least 1 chronic disease, asthma was the most commonly reported among people in the 0- to 21-year age group (31 [16.5%]), whereas hypertension was the most reported for the 22- to 65-year age group (99 [26.1%]) and the older than 65 years group (29 [59.2%]). The study addressed the need to consider chronic conditions and ensure primary care accessibility for vulnerable populations during recovery from a disaster.

Ridenour et alReference Ridenour, Cummings, Sinclair and Bixler14 conducted a needs assessment to identify the needs of evacuees in West Virginia after Hurricane Katrina. A health status questionnaire adapted from a CDC surveillance instrument and West Virginia University School of Medicine at Morgantown medical screening tool was used to assess acute conditions, chronic medical conditions, and current needs. Surveys were linked to Red Cross household registration records that included information on the following: (1) address before the disaster, (2) dwelling type and homeowner insurance, (3) total household income, (4) incurred damage from the hurricane on the home, and (5) current housing needs. Only 51% of evacuees responded to the survey. A quarter of respondents had an acute condition, and 46% had at least 1 chronic medical condition at the time of the survey. Current medical needs expressed by the evacuees included the following: dental care, 57%; eyeglasses, 34%; dentures, 28%; and medical services, 25%. The study noted that chronic disease management and medical equipment needs are often overlooked priorities among displaced populations after a disaster.

Following a cursory examination of publications in the lay press, we found hundreds of newsprint articles addressing our proposed search categories (Table 3). For example, the search terms “health disparities and disasters” resulted in 996 hits with articles. On further searching, we noted that these articles dated back to 1982. These findings differed greatly from the number of health studies found in the literature for this same category, wherein applicable studies did not emerge until around 2005. Results from newsprint hits signify a more relevant and engaged media.

TABLE 3

Summary of Newsprint Hits in Lexis-Nexis From January 1, 1970, to May 15, 2009

DISCUSSION

A dearth of research exists on the effects of a disaster for communities disproportionately affected by health and health care disparities. Few published studies assess a disaster's effect on chronic disease mortality and morbidity among medically underserved populations. Contrasting sharply with these glaring deficits apparent in the literature are countless articles in the lay press that have more than adequately addressed the issue of underlying health disparities in the wake of a disaster. For example, a seminal article was featured in the Washington Post just two weeks after Hurricane Katrina. The article, “At Risk Before the Storm Struck: Prior Health Disparities Due to Race, Poverty Multiply Death, Disease,” stressed that many of the disaster-affected areas had “a bunch of people who have less thanoptimal healthcare to begin with, and they have a large number of these diseases that people who get less than optimal healthcare end up getting” that in turn “left this high-risk group in greater peril than those with better health and access to care.”Reference Fowkes, Blossom, Anderson and Sandrock72

The lay press has been at the forefront of this major public health issue, with thousands of newsprint articles discussing this issue now in circulation, while disaster epidemiology and health and health care disparities research have largely been out of touch, as seen by the paucity of peer-reviewed literature. Although it is difficult to say why, we could only speculate that the media's ability to identify this issue as “newsworthy” suggests that it is much more conscientious and responsive to issues faced by its audience. People will continue to suffer from the effects of disasters. As a result, public health professionals must become more deliberate at accomplishing research directives that tackle real-world issues.

CONCLUSIONS

The precise mechanisms through which a community with preexisting health and health care disparities may become more susceptible to the deleterious effects of a disaster are unknown. The focus on the context of health and health care disparities before and after a disaster is important because the current literature does not capture the subsequent strain on medical resources imposed by the secondary surge of incident chronic disease in medically underserved communities. Given that chronic disease accounts for the greatest cause of morbidity in the United States, especially in rural areas, it is imperative to address chronic disease management in disaster response and identify ways to mitigate the health and health care disparities associated with populations vulnerable to disaster.

To better facilitate disaster recovery, it is vital that study designs account for effect modification by health and health care disparities for more accurate risk assessment models. Future studies should account for differences in health status and access before and after the disaster to better address the interaction between disparities in health and health care and adverse health outcomes in disaster populations. Disaster preparedness is not often a leading priority for most communities in the United States. Furthermore, lessons learned from Hurricane Katrina regarding existent health and health care disparities and increased morbidity and mortality among poor populations should be more extensively documented so that they can be used to plan for future disaster events.

The Robert T. Stafford Disaster Relief and Emergency Assistance Act (the Stafford Act)Reference Robert153 governs the federal response to disasters within the United States, including provisions for medical care and treatment of injured victims. However, the act does not address health and health care disparities or the specific needs of communities that are medically underserved. Therefore, further research is needed to better understand the most appropriate approaches to addressing health and health care disparities in the context of the Stafford Act. Factors contributing to health and health care disparities must be accounted for in disaster planning and response to ensure that rural and vulnerable populations are equipped to be resilient during the initial and secondary surges and other disaster-related events. We believe that the secondary surge of incident chronic disease after a disaster is an underappreciated phenomenon and needs further study.