Salmonellosis from inadequately pasteurized milk--Kentucky.

In late April 1 984, three isolates of Salmonella typhimurium, all from specimens from per­ sons associated with a convent in western Kentucky, were reported by the Kentucky Division for Laboratory Services to the Division of Epidemiology. Subsequent investigation revealed that at least 16 cases of gastroenteritis (predominantly diarrhea) were associated with the convent between March 28 and May 2. The likely vehicle was inadequately pasteurized milk. One hundred forty nuns reside at the convent; additional persons are employed as caretak­ ers, foodhandlers, and farm workers. All meals are prepared in the convent kitchen and are available to residents and employees. In early May, 1 80 (90%) of 200 persons at risk filled out questionnaires on basic demo­ graphic information, recent gastrointestinal illness, and frequency of consuming milk, raw milk, chicken, turkey, pork, and eggs. Sixteen (9%) persons met the case definition of a posi­ tive stool culture for S. typhimurium or at least three loose stools per day lasting 2 or more days or any three of the following symptoms, fever, diarrhea (not meeting the criteria above), nausea and/or vomiting, or abdominal cramps; 1 2 additional persons (7%) who had gastroin­ testinal symptoms but did not meet the case definition were excluded from the statistical analysis. Ill persons had diarrhea (100%), abdominal cramps (63%), nausea (50%), fever (44%), and vomiting (13%). Diarrhea lasted 1-8 days (median 3 days), with two to 11 loose stools per day (median five). Patients were 21 years to 86 years of age (median 67 years). Fifteen (94%) were female, and 14 (88%) were convent residents. Three (1 9%) of the 16 consulted a physician. Of persons filling out questionnaires, 14 (15%) of 91 persons who admitted drinking pas­ teurized milk became ill, but only two (3%) of 75 who claimed not to have drunk milk became ill. Persons drinking pasteurized milk were approximately six times more likely to develop ill­ ness (p = 0.01). No other risk factors were identified. Fourteen (88%) of the 16 ill persons had onset on or after March 28, with three clusters of cases approximately 2 weeks apart. In late April, 24 symptomatic nuns and 1 8 asymptomatic foodhandlers submitted single stool-culture specimens. Five (31%) of the 16 ill persons, all nuns, had stool cultures positive for S. typhimurium; eight (50%) were negative; and three (19%) had no stool specimen submitted. A 69-year-old culture-positive ill nun was hospital­ ized for a Guillain-Barre-like illness 20 days after onset of gastrointestinal symptoms. A raw milk sample collected May 9 also yielded S. typhimurium. The isolate from milk and three iso­ lates from humans had identical plasmid profiles. The remaining two human isolates were reported to be identical to each other and were probably equivalent to the other isolates. Anti­ biograms of all six isolates were the same. Before early March 1984, the convent had its own herd of dairy cattle and pasteurized its own milk. At that time, the convent began purchasing raw milk from a Grade-A dairy farm in