OBJECTIVE To determine the cause of a community-wide outbreak of cryptosporidiosis. METHODS A matched case-control study. METHODS General community of Lane County, Oregon. METHODS Persons with Cryptosporidium detected in their stool from June to October 1992 were identified by contacting laboratories serving the area. Exposures of the first 18 case patients identified were compared with those of 18 age- and neighborhood-matched controls selected from a reverse telephone directory. METHODS Reported exposures to risk factors for cryptosporidiosis and abatement of cryptosporidiosis outbreak. RESULTS Fifty-five patients with cryptosporidiosis were detected, including 37 who were the first individuals ill in their households. The case-control study involving the first 18 case patients showed no association between illness and attendance at day care or drinking municipal water or drinking untreated surface waters (river or lake water) in the 2 weeks before onset of illness. However, nine of 18 case patients reported swimming at a local wave pool, compared with none of 18 controls. We ultimately identified 17 case patients who reported swimming at the same wave pool during their incubation periods, whose exposure dates spanned a 2-month period. Inspection of the pool's filtration system did not detect any abnormalities. The outbreak subsided after the pool water was drained and replaced. CONCLUSIONS This prolonged outbreak of cryptosporidiosis was likely caused by exposure to fecally contaminated wave pool water. Since Cryptosporidium is highly chlorine resistant and inadequately removed by sand filters, such outbreaks may represent an unrecognized hazard of wave pools, where the likelihood of inadvertent water ingestion is high. Such outbreaks may go undetected in areas where cryptosporidiosis is not reportable or laboratory screening is infrequent.