A PROSPECTIVE STUDY OF SWIMMING-RELATED ILLNESS AT FOUR POPULAR BEACHES IN TRINIDAD, WEST INDIES

Abstract

ISEE-539 Abstract: In 1992, an epidemiological/water quality study was conducted at four beaches in Trinidad during both the dry and wet seasons, to assess swimming-associated health effects. A total of 10, 204 persons were interviewed at the selected beaches. Follow-up interviews were conducted by telephone, three days after the initial beach interview. Bacterial indicators of water quality used were total coliforms, faecal coliforms, Escherichia coli and enterococci. There was seasonal variation in water quality at the four beaches with higher bacteria levels in the wet season than in the dry. Only two of the six survey areas at the four beaches had bacteria levels in compliance with the U.S. EPA 1976 and 1986 bathing beach standards. Illnesses investigated were respiratory illness, gastrointestinal illness, highly credible gastrointestinal illness, allergies, eye and ear infections and “other illness” (fever, headache or any other symptom). Rates varied significantly (p<0.001) among the various illnesses. Symptoms of respiratory illness were reported most frequently with an illness rate of 39 persons/1000 persons. There was a significantly higher proportion of participants with water contact reporting one or more symptoms (p<0.001), symptoms of ear infections (p<0.05), eye infections (p<0.05), allergies (p<0.01), skin infections (p<0.01) and gastrointestinal illness (p<0.05) than those without water contact. Illness rates also varied with season and survey area as well as with age and gender of study participants. A significantly higher (p<0.01) proportion of participants reported symptoms of ear infection and highly credible gastrointestinal illness in the dry season than in the wet season. Among the survey areas, there were significant differences in illness rates for one or more symptoms (p<0.001), symptoms of highly credible gastrointestinal illness (p<0.001), allergies (p<0.001), skin infections (p<0.001) and respiratory illness (p<0.01). Younger participants reported higher illness rates for most of the illnesses investigated. Highly credible gastrointestinal illness rates were significantly different (p<0.001) among the different age groups in the study population. Females had higher illness rates than males for most illnesses. Multiple regression models were fitted to examine the effects of various water quality parameters as predictors of illness rates. The models showed that while some water quality parameters gave an indication of illness rates, most of the models fitted did not have high adjusted r2 values.