RESUMO
OBJECTIVES: On January 9, 2014, approximately 10 000 gallons of a mixture of 4-methylcyclohexanemethanol and propylene glycol phenyl ether spilled into West Virginia's Elk River, contaminating the potable water supply of about 300 000 West Virginia residents. This study sought to describe acute health effects after the chemical spill. METHODS: We conducted a descriptive analysis using 3 complementary data sources: (1) medical records of patients who visited an emergency department during January 9-23, 2014, with illness potentially related to the spill; (2) West Virginia Poison Center caller records coded as "contaminated water" during January 9-23, 2014; and (3) answers to household surveys about health effects from a Community Assessment for Public Health Emergency Response (CASPER) questionnaire administered 3 months after the spill. RESULTS: In the 2 weeks after the spill, 2000 people called the poison center reporting exposure to contaminated water, and 369 people visited emergency departments in the affected area with reports of exposure and symptoms potentially related to the spill. According to CASPER weighted cluster analyses, an estimated 25 623 households (21.7%; 95% confidence interval [CI], 14.4%-28.9%) had ≥1 person with symptoms who felt that they were related to the spill in the 3 months after it. Reported health effects across all 3 data sources included mild skin, respiratory, and gastrointestinal symptoms that resolved with no or minimal treatment. CONCLUSIONS: Medical records, poison center data, and CASPER household surveys were inexact but useful data sources to describe overall community health effects after a large-scale chemical spill. Analyzing multiple data sources could inform epidemiologic investigations of similar events.