American Journal of Epidemiology Copyright © 1992 by The Johns Hopkins University School of Hygiene and Pubfc Health Al rights reserved
Vol. 135, No 12 Printed in U.S A.
A Trichinosis Outbreak among Southeast Asian Refugees
The number of cases of trichinosis reported to Centers for Disease Control has declined steadily, with an average of only 44 cases per year from 1984 through 1988. This decline was almost entirely due to a reduction in cases acquired from ingestion of fresh commercial pork. However, from July 21 through September 3, 1990, 90 (72%) of 125 Southeast Asian refugees from six states and Canada developed trichinosis after attending or eating pork sausage taken from a wedding held in Des Moines, Iowa, on July 14, 1990. Eating uncooked sausage prepared at home from commercially obtained pork was associated with the development of this illness (odds ratio = 34.0, p < 0.001). Analysis by amount of pork consumed was significant (Mann-Whitney U rank sum test, p < 0.001). This outbreak of trichinosis in Iowa is the fourth reported within the last 15 years among the 900,000 Southeast Asian refugees resident in the United States and one of the largest reported outbreaks in US history. The continued presence of Trichinella spiralis in commercial pork emphasizes the need for further education and control measures for persons whose dietary habits place them at risk for developing trichinosis. Am J Epidemiol 1992;135:1404-10. Asian Americans; Trichinella; trichinosis
Since 1975, more than 900,000 Southeast Asian refugees have entered the United States (1). Recent studies have shown that this population is at increased risk of developing trichinosis after they have settled in this country (2). Three outbreaks of trichinosis related to the consumption of underReceived for publication September 4, 1991, and in final form January 6, 1992. 1 Parasitic Diseases Branch, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control, Public Health Service, US Department of Health and Human Services, Atlanta, GA. 2 Iowa Department of Public Health, Des Moines, IA. 3 Current address' Rush Presbyterian-St Luke's Medical Center, Chicago, IL. Reprint requests to Dr. Peter M. Schantz, Chamblee 23/F-13, Parasitic Diseases Branch, Centers for Disease Control, 1600 Clifton Road, Atlanta, GA 30333. The authors thank an who helped with this investigation, including the staff of the Iowa Public Health Department. They especially thank Vannavong Phabmixay, who served as translator and provided invaluable insight into Laotian culture in the Des Moines area. Finally, they are grateful to the laboratory staff who performed the serotogic testing, Mananna Wflson and Dons Ware.
cooked pork purchased directly from small farms have been previously reported in Southeast Asians (2). The Bureau of Refugee Programs in Des Moines estimates that there are 10,500 Southeast Asian refugees in Iowa. The 2,000 Laotians and 1,500 Thaidam who live in the Des Moines area participate in frequent weddings and parties during the summer and fall months. Thaidam are a people who are linguistically and culturally similar to Laotians, but who lived in Vietnam. Most members of both communities attend several of these functions. Trichinosis has been reported only sporadically in people who have eaten wild game in Laos and Vietnam (3), and there have been no reported cases in Kampuchea (3). Thailand has reported 3,000 cases since 1962, predominantly in the northern hill tribe people (4). In general, trichinosis is not considered a significant health problem in Southeast Asia, even though uncooked pork
1404
Downloaded from https://academic.oup.com/aje/article-abstract/135/12/1404/88774 by University of Western Sydney user on 13 January 2019
James B. McAuley,13 Marco K. Michelson,1 Allen W. Hightower,1 Susan Engeran,1 Laverne A. Wintermeyer,2 and Peter M. Schantz1
Trichinosis in Southeast Asian Refugees
MATERIALS AND METHODS
Preliminary evaluation indicated that many of the ill persons had eaten uncooked pork at the wedding on July 14. However, several persons also indicated that they had eaten undercooked pork at parties on July 21 and July 28. A case-control study was initiated to determine which of the three parties was the source of the Trichinella 5p/ra//5-infected meat and to identify possible risk factors associated with developing trichinosis. Since an increase in febrile illnesses was reportedly occurring in the community during the trichinosis outbreak, the Centers for Disease Control surveillance case definition for trichinosis was modified to provide a more specific indicator of clinical disease than would have been possible using the existing definition (5). A case was defined as serologic or muscle biopsy-proven T. spiralis infection and clinical illness occurring in a person after July 14 and consisting of at least two of the following: fever, myalgia, or periorbital edema, plus any one of the following: eosinophilia, weakness, headache, nausea, vomiting, diarrhea, or abdominal pain. When a person met criterion 1, cases associated with him or her were identified by either a positive serologic test for trichinosis or clinical illness, as defined above, after July 14 occurring in individuals who had attended or eaten food taken from any one of
the three parties indicated by the initial patients as likely places of exposure. For the case-control study, ill persons originally identified by the Iowa Department of Public Health were asked if they would be willing to answer a detailed questionnaire and submit blood specimens for serologic testing. In addition, they were asked to identify any other family members or friends who had not been ill recently and who might participate in the study as controls. Only persons who submitted blood for serologic testing were included in the case-control study. Persons who met the case definition as an associated case but did not have serologic confirmation were included only in the determination of the total number of cases of trichinosis. After the wedding party on July 14 was statistically implicated, odds ratios for the remaining variables were calculated using only controls who had attended or eaten food taken from that party. As no official guest list was available, further case finding was performed by contacting persons identified by an employee of the Iowa Refugee Health Program who had also attended the wedding. Out-of-town guests were interviewed by telephone, but were not included in the case-control analysis because serologic confirmation could not be obtained. Physicians identified by those interviewed and physicians known to care frequently for Southeast Asian patients were asked if they had encountered any cases of trichinosis in the previous 2 months. Data were collected on basic demographic characteristics, clinical presentation, and several potential risk factors for developing trichinosis. All blood specimens were sent to the Centers for Disease Control Parasitic Diseases Laboratory, where serology was performed using the Bentonite flocculation test (6). Data were analyzed using the statistical computer package, Epi Info version 5. Statistical differences in proportions or prevalences between cases and controls were determined using the chi-square and Fisher's exact tests, and precision was expressed using Cornfield 95% confidence intervals. Length of residence in the United States and
Downloaded from https://academic.oup.com/aje/article-abstract/135/12/1404/88774 by University of Western Sydney user on 13 January 2019
is a traditional part of the diet in this region. Whether this phenomenon is due to the absence or underdiagnosis of infection is not known. On August 14, 1990, the Iowa Department of Public Health was notified by three different physicians of five probable cases of trichinosis in Laotian immigrants who had attended a wedding in Des Moines on July 14, 1990. Investigation by the Department of Health and the Centers for Disease Control ultimately identified an interstate and international outbreak among the approximately 250 persons exposed to uncooked pork sausage served at the wedding.
1405
1406
McAuley et a).
ding and were included in the case-control study (n = 54), analysis of the food items eaten showed a significant association between eating pork sausage and acquiring infection (OR = 34.0, 95 percent CI 4.1373.0, p < 0.001); consumption of other food items was not associated with the development of trichinosis (table 2). The more pork a person ate, the more likely he or she was to become ill, although the number of people who ate more than three servings was small (table 3) (p < 0.001, Mann-Whitney U rank sum test). No significant differences in age or ethnic group was observed for cases compared with controls. In response to questions about general consumption of uncooked pork, four (10 percent) of 40 cases reported eating undercooked pork more than once a month compared with four (44 percent) of nine controls (OR = 0.24, 95 percent CI 0.04-1.49, p = 0.07). No difference in the usual source of pork was noted between cases and controls. There was no association between infection and length of residence in the United States (p > 0.05,
RESULTS Case-control study
Sixty-nine persons participated in the case-control study; 41 participants met the case definition, and 28 participants served as controls. Thirteen controls attended the July 14 wedding and 15 did not. Three people initially thought to be controls were changed to cases after their serology results were found to be strongly positive. The risk of acquiring infection for each of the parties held in July is listed in table 1. Only attendance at the July 14 wedding was associated with a significant odds ratio (odds ratio (OR) = 22.5, 95 percent confidence interval (CI) 4.0-167.9, p < 0.001). Among the people who attended the wed-
TABLE 1. Association between Illness and party attended, Iowa, 1990 Attended or ate food ( " • • 69)
Evert
Date
Patients Yes
Wedding Party Wedding Party Party Other
7/14 7/14 7/21 7/21 7/28
Controls No
Yes
39
2
13
15
0 9 3 4
41 32 38 37 40
1 4 2
27 24 26 28 27
1
OR*
95% CI*
p valuef
22.5
4.0-167.9
Vol. 135, No 12 Printed in U.S A.
A Trichinosis Outbreak among Southeast Asian Refugees
The number of cases of trichinosis reported to Centers for Disease Control has declined steadily, with an average of only 44 cases per year from 1984 through 1988. This decline was almost entirely due to a reduction in cases acquired from ingestion of fresh commercial pork. However, from July 21 through September 3, 1990, 90 (72%) of 125 Southeast Asian refugees from six states and Canada developed trichinosis after attending or eating pork sausage taken from a wedding held in Des Moines, Iowa, on July 14, 1990. Eating uncooked sausage prepared at home from commercially obtained pork was associated with the development of this illness (odds ratio = 34.0, p < 0.001). Analysis by amount of pork consumed was significant (Mann-Whitney U rank sum test, p < 0.001). This outbreak of trichinosis in Iowa is the fourth reported within the last 15 years among the 900,000 Southeast Asian refugees resident in the United States and one of the largest reported outbreaks in US history. The continued presence of Trichinella spiralis in commercial pork emphasizes the need for further education and control measures for persons whose dietary habits place them at risk for developing trichinosis. Am J Epidemiol 1992;135:1404-10. Asian Americans; Trichinella; trichinosis
Since 1975, more than 900,000 Southeast Asian refugees have entered the United States (1). Recent studies have shown that this population is at increased risk of developing trichinosis after they have settled in this country (2). Three outbreaks of trichinosis related to the consumption of underReceived for publication September 4, 1991, and in final form January 6, 1992. 1 Parasitic Diseases Branch, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control, Public Health Service, US Department of Health and Human Services, Atlanta, GA. 2 Iowa Department of Public Health, Des Moines, IA. 3 Current address' Rush Presbyterian-St Luke's Medical Center, Chicago, IL. Reprint requests to Dr. Peter M. Schantz, Chamblee 23/F-13, Parasitic Diseases Branch, Centers for Disease Control, 1600 Clifton Road, Atlanta, GA 30333. The authors thank an who helped with this investigation, including the staff of the Iowa Public Health Department. They especially thank Vannavong Phabmixay, who served as translator and provided invaluable insight into Laotian culture in the Des Moines area. Finally, they are grateful to the laboratory staff who performed the serotogic testing, Mananna Wflson and Dons Ware.
cooked pork purchased directly from small farms have been previously reported in Southeast Asians (2). The Bureau of Refugee Programs in Des Moines estimates that there are 10,500 Southeast Asian refugees in Iowa. The 2,000 Laotians and 1,500 Thaidam who live in the Des Moines area participate in frequent weddings and parties during the summer and fall months. Thaidam are a people who are linguistically and culturally similar to Laotians, but who lived in Vietnam. Most members of both communities attend several of these functions. Trichinosis has been reported only sporadically in people who have eaten wild game in Laos and Vietnam (3), and there have been no reported cases in Kampuchea (3). Thailand has reported 3,000 cases since 1962, predominantly in the northern hill tribe people (4). In general, trichinosis is not considered a significant health problem in Southeast Asia, even though uncooked pork
1404
Downloaded from https://academic.oup.com/aje/article-abstract/135/12/1404/88774 by University of Western Sydney user on 13 January 2019
James B. McAuley,13 Marco K. Michelson,1 Allen W. Hightower,1 Susan Engeran,1 Laverne A. Wintermeyer,2 and Peter M. Schantz1
Trichinosis in Southeast Asian Refugees
MATERIALS AND METHODS
Preliminary evaluation indicated that many of the ill persons had eaten uncooked pork at the wedding on July 14. However, several persons also indicated that they had eaten undercooked pork at parties on July 21 and July 28. A case-control study was initiated to determine which of the three parties was the source of the Trichinella 5p/ra//5-infected meat and to identify possible risk factors associated with developing trichinosis. Since an increase in febrile illnesses was reportedly occurring in the community during the trichinosis outbreak, the Centers for Disease Control surveillance case definition for trichinosis was modified to provide a more specific indicator of clinical disease than would have been possible using the existing definition (5). A case was defined as serologic or muscle biopsy-proven T. spiralis infection and clinical illness occurring in a person after July 14 and consisting of at least two of the following: fever, myalgia, or periorbital edema, plus any one of the following: eosinophilia, weakness, headache, nausea, vomiting, diarrhea, or abdominal pain. When a person met criterion 1, cases associated with him or her were identified by either a positive serologic test for trichinosis or clinical illness, as defined above, after July 14 occurring in individuals who had attended or eaten food taken from any one of
the three parties indicated by the initial patients as likely places of exposure. For the case-control study, ill persons originally identified by the Iowa Department of Public Health were asked if they would be willing to answer a detailed questionnaire and submit blood specimens for serologic testing. In addition, they were asked to identify any other family members or friends who had not been ill recently and who might participate in the study as controls. Only persons who submitted blood for serologic testing were included in the case-control study. Persons who met the case definition as an associated case but did not have serologic confirmation were included only in the determination of the total number of cases of trichinosis. After the wedding party on July 14 was statistically implicated, odds ratios for the remaining variables were calculated using only controls who had attended or eaten food taken from that party. As no official guest list was available, further case finding was performed by contacting persons identified by an employee of the Iowa Refugee Health Program who had also attended the wedding. Out-of-town guests were interviewed by telephone, but were not included in the case-control analysis because serologic confirmation could not be obtained. Physicians identified by those interviewed and physicians known to care frequently for Southeast Asian patients were asked if they had encountered any cases of trichinosis in the previous 2 months. Data were collected on basic demographic characteristics, clinical presentation, and several potential risk factors for developing trichinosis. All blood specimens were sent to the Centers for Disease Control Parasitic Diseases Laboratory, where serology was performed using the Bentonite flocculation test (6). Data were analyzed using the statistical computer package, Epi Info version 5. Statistical differences in proportions or prevalences between cases and controls were determined using the chi-square and Fisher's exact tests, and precision was expressed using Cornfield 95% confidence intervals. Length of residence in the United States and
Downloaded from https://academic.oup.com/aje/article-abstract/135/12/1404/88774 by University of Western Sydney user on 13 January 2019
is a traditional part of the diet in this region. Whether this phenomenon is due to the absence or underdiagnosis of infection is not known. On August 14, 1990, the Iowa Department of Public Health was notified by three different physicians of five probable cases of trichinosis in Laotian immigrants who had attended a wedding in Des Moines on July 14, 1990. Investigation by the Department of Health and the Centers for Disease Control ultimately identified an interstate and international outbreak among the approximately 250 persons exposed to uncooked pork sausage served at the wedding.
1405
1406
McAuley et a).
ding and were included in the case-control study (n = 54), analysis of the food items eaten showed a significant association between eating pork sausage and acquiring infection (OR = 34.0, 95 percent CI 4.1373.0, p < 0.001); consumption of other food items was not associated with the development of trichinosis (table 2). The more pork a person ate, the more likely he or she was to become ill, although the number of people who ate more than three servings was small (table 3) (p < 0.001, Mann-Whitney U rank sum test). No significant differences in age or ethnic group was observed for cases compared with controls. In response to questions about general consumption of uncooked pork, four (10 percent) of 40 cases reported eating undercooked pork more than once a month compared with four (44 percent) of nine controls (OR = 0.24, 95 percent CI 0.04-1.49, p = 0.07). No difference in the usual source of pork was noted between cases and controls. There was no association between infection and length of residence in the United States (p > 0.05,
RESULTS Case-control study
Sixty-nine persons participated in the case-control study; 41 participants met the case definition, and 28 participants served as controls. Thirteen controls attended the July 14 wedding and 15 did not. Three people initially thought to be controls were changed to cases after their serology results were found to be strongly positive. The risk of acquiring infection for each of the parties held in July is listed in table 1. Only attendance at the July 14 wedding was associated with a significant odds ratio (odds ratio (OR) = 22.5, 95 percent confidence interval (CI) 4.0-167.9, p < 0.001). Among the people who attended the wed-
TABLE 1. Association between Illness and party attended, Iowa, 1990 Attended or ate food ( " • • 69)
Evert
Date
Patients Yes
Wedding Party Wedding Party Party Other
7/14 7/14 7/21 7/21 7/28
Controls No
Yes
39
2
13
15
0 9 3 4
41 32 38 37 40
1 4 2
27 24 26 28 27
1
OR*
95% CI*
p valuef
22.5
4.0-167.9
