Surveillance of shelters after Hurricane Hugo

Surveillance of shelters after Hurricane Hugo – Puerto Rico

Surveillance of Shelters after Hurricane Hugo — Puerto Rico

On September 18, 1989, Hurricane Hugo struck the northeastern coast of Puerto Rico and left [greater than] 75% of the island without electricity or water[1]. An estimated 10,300 persons were displaced from their residences to temporary shelters. Because of the potential for outbreaks of infectious diseases (such as gastroenteritis) among shelter residents, the Puerto Rico Department of Health (PRDH) conducted active public health surveillance of the shelters from September 25 through October 3. The surveillance system was designed to 1) obtain information about the occurrence of infectious diseases; 2) determine the availability of shelter facilities such as water, toilets, electricity, and medical care; 3) monitor sanitation; 4) identify health-related topics about which to educate shelter residents; and 5) assist in resource allocation.

The PRDH divides Puerto Rico into eight administrative regions, six of which were affected by the hurricane. For this surveillance system, each region was assigned teams of public health professionals who reported daily to a supervisory regional epidemiologist, a regional medical director, and the PRDH Division of Epidemiology (DOE). Each team inspected eight shelters daily to provide health education and to assess overcrowding, the occurrence of infectious diseases, and the availability of water and toilet facilities. Potential infectious disease outbreaks were investigated by the regional epidemiologist, with assistance from the PRDH DOE. An infectious disease outbreak was defined as three or more cases of any disease reported on any day in one shelter or an attack rate [greater than] 10% of a shelter’s average population during a 5-day period. The director of DOE reviewed the teams’ reports and reported to the Secretary of Health daily.

An estimated 10,300 persons were housed in 161 temporary shelters; 113 (70%) shelters were in public schools, 38 (24%) in community centers, and 10 (6%) in churches. The number of persons housed per shelter ranged from five to 297 (mean: 64). The surveillance system monitored 158 shelters. On September 25, 9429 persons were housed in the shelters — 20 (13%) of which lacked running water; on October 3, 7528 persons were housed in the shelters — all of which had running water. Three shelters were not monitored — one on the island of Culebra housing [greater than] 100 persons and two on the main island housing [less than] 10 persons each.

From September 25 through September 29, five shelters reported a [greater than] 10% incidence of gastroenteritis. However, investigation failed to confirm these outbreaks; reported information either could not be verified or potential cases did not meet the PRDH gastroenteritis case definition (i.e., three or more loose stools within a 24-hour period).

Outbreaks of head lice infestation were confirmed in 28 (18%) shelters; at least one case of lice was reported in each of 34 (22%) shelters. Outbreaks of influenza-like illness occurred in 18 (11%) shelters. Compared with persons housed in schools, those housed in community centers and churches were 4.2 and 3.4 times more likely, respectively, to have had an influenza-like illness.

Editorial Note: Before impending hurricanes and other potential disasters, many persons are reluctant to evacuate to shelters. In Puerto Rico, evacuation posed few problems, and the benefits of evacuation far outweighed the risks of residing in shelters. (The three persons who died during the impact phase of Hurricane Hugo had repeatedly refused to evacuate to shelters [1; CDC, unpublished data]). The PRDH surveillance did not detect or document serious disease outbreaks or other public health problems in the shelters; moreover, only a few minor health problems were identified in the shelters.

The difference in risk for certain infectious diseases between persons housed in schools and persons housed in community centers and churches was not related to the number of persons per shelter or to overcrowding. One possible explanation for this difference was the use of multiple small rooms for housing persons at schools rather than the use of one large room at community centers and churches. The head lice infestation may have reflected endemic occurrence detected only incidentally by this surveillance system.

The PRDH shelter surveillance system resulted in several public health interventions: provision of portable toilets to shelters that needed them; delivery of potable water to shelters that lacked drinking water; education of shelter residents about personal hygiene; and treatment of persons with head lice infestations. In addition, data from the surveillance system were useful in assessing and dispelling rumors that circulated the day after the hurricane about outbreaks of infectious diseases (including cholera) in the shelters, as well as ensuring that basic sanitary services and potable water were provided efficiently. Public health surveillance in large shelters, such as that established by the PRDH after Hurricane Hugo and after torrential rains in 1985[2], should be part of the public health emergency response to disasters.


[1]CDC. Deaths associated with Hurricane Hugo — Puerto Rico. MMWR 1989;38:680-2. [2]Dietz VJ, Sanderson L, Diaz L, Rigau-Perez JG. Morbidity/mortality surveillance after torrential rains, Puerto Rico [Abstract]. In: Program and abstracts of the Epidemic Intelligence Service 35th Annual Conference. Atlanta: US Department of Health and Human Services, Public Health Service, CDC, 1986:84.

COPYRIGHT 1990 U.S. Government Printing Office

COPYRIGHT 2004 Gale Group