Outbreak of Salmonellosis Linked to Live Poultry from a Mail-Order Hatchery

Transcription

1 original article Outbreak of Salmonellosis Linked to Live Poultry from a Mail-Order Hatchery Nicholas H. Gaffga, M.D., M.P.H., Casey Barton Behravesh, D.V.M., Dr.P.H., Paul J. Ettestad, D.V.M., Chad B. Smelser, M.D., Andrew R. Rhorer, M.S., Alicia B. Cronquist, R.N., M.P.H., Nicole A. Comstock, M.S.P.H., Sally A. Bidol, M.P.H., Nehal J. Patel, M.P.H., Peter Gerner-Smidt, M.D., D.Med.Sci., William E. Keene, Ph.D., M.P.H., Thomas M. Gomez, D.V.M., Brett A. Hopkins, D.V.M., Ph.D., Mark J. Sotir, Ph.D., M.P.H., and Frederick J. Angulo, D.V.M., Ph.D. A bs tr ac t Background Outbreaks of human salmonella infections are increasingly associated with contact with live poultry, but effective control measures are elusive. In 005, a cluster of human salmonella Montevideo infections with a rare pattern on pulsed-field gel electrophoresis (the outbreak strain) was identified by PulseNet, a national subtyping network. Methods In cooperation with public health and animal health agencies, we conducted multistate investigations involving patient interviews, trace-back investigations, and environmental testing at a mail-order hatchery linked to the outbreak in order to identify the source of infections and prevent additional illnesses. A case was defined as an infection with the outbreak strain between 004 and 0. Results From 004 through 0, we identified 36 cases in 43 states. The median age of the patient was 4 years. Interviews were completed with 56 patients (or their caretakers) (49%), and 36 of these patients (3%) were hospitalized. Among the 45 patients for whom information was available, 80 (55%) had bloody diarrhea. Information on contact with live young poultry was available for 59 patients, and of these patients (77%) reported having such contact. A mail-order hatchery in the western United States was identified in 8% of the trace-back investigations, and the outbreak strain was isolated from samples collected at the hatchery. After interventions at the hatchery, the number of human infections declined, but transmission continued. From the Division of Foodborne, Waterborne, and Environmental Diseases, National Center for Emerging and Zoonotic Infectious Diseases, Centers for Disease Control and Prevention, Atlanta (N.H.G., C.B.B., N.J.P., P.G.-S., M.J.S., F.J.A.); the New Mexico Department of Health, Santa Fe (P.J.E., C.B.S.); the Animal and Plant Health Inspection Service, National Poultry Improvement Plan (A.R.R.) and Veterinary Services (T.M.G.), Department of Agriculture, Washington, DC; the Colorado Department of Public Health and Environment, Denver (A.B.C., N.A.C.); the Michigan Department of Community Health, Lansing (S.A.B.); the Oregon Public Health Division, Portland (W.E.K.); and Pfizer Animal Health, Overland Park, KS (B.A.H.). Address reprint requests to Dr. Gaffga at the Centers for Disease Control and Prevention, 600 Clifton Rd., NE MS E-0, Atlanta, GA 30333, or at ngaffga@cdc.gov. N Engl J Med 0;366: Copyright 0 Massachusetts Medical Society. Conclusions We identified a prolonged multistate outbreak of salmonellosis, predominantly affecting young children and associated with contact with live young poultry from a mail-order hatchery. Interventions performed at the hatchery reduced, but did not eliminate, associated human infections, demonstrating the difficulty of eliminating salmonella transmission from live poultry. n engl j med 366; nejm.org may 3, Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

2 Nontyphoidal salmonella infections are associated with substantial morbidity and mortality in the United States, with an estimated million illnesses, 9,000 hospitalizations, and 370 deaths occurring annually. Although the majority of these infections are foodborne, salmonellosis can also be acquired through contact with animals, including live poultry (e.g., chickens, ducks, geese, and turkeys). -8 Poultry can become infected with salmonella through comingling with infected birds from different sources, vertical transmission from infected hens, or contaminated feed. 9 Live poultry infected with salmonella typically appear healthy but can shed bacteria intermittently, making the sampling of individual birds an unreliable way to determine whether birds are carrying or shedding bacteria (unpublished data). Poultry vaccines for salmonella can reduce bacterial burdens in flocks but have not been proved effective in eliminating bacteria from hatchery environments. In the United States, approximately 50 million live poultry are sold by mail-order hatcheries annually, generating between $50 million and $70 million in sales. In 009, a total of 0 mail-order hatcheries sold poultry (unpublished data). Although the mail-order hatchery industry is smaller than the commercial poultry and egg industry, which generates billions of dollars a year, hatcheries in the two industries are encouraged to maintain the same standards of environmental hygiene (unpublished data). Guidance on achieving these standards is provided by the National Poultry Improvement Plan (NPIP), at the U.S. Department of Agriculture (USDA); however, compliance with recommendations by the owners of mail-order hatcheries is voluntary. Within 4 hours after hatching, live young poultry (e.g., chicks, goslings, and ducklings) are frequently shipped in cardboard boxes containing up to 00 birds through the U.S. Postal Service from hatcheries to agricultural feed stores or private homes around the country. These birds can be purchased for $5 or less in venues such as agricultural feed stores or directly from mail-order hatcheries. The Postal Service reported that in the first 6 months of 009, it shipped approximately 544,300 kg (. million lb) of packages containing live poultry; during this same period, the mail-order hatchery industry reported record sales due to increased interest in raising backyard flocks and urban chickens. 0 Traceback investigations of outbreaks associated with hatcheries are complicated by a practice called drop-shipping, in which a hatchery that cannot fill a customer s order requests that a second hatchery send birds directly to the customer under the name of the first hatchery. In these situations, the receipt does not indicate the name of the second hatchery, complicating the identification of the hatchery of origin for a particular shipment of birds. This report describes the investigation of an ongoing multistate outbreak of human salmonella Montevideo infections associated with contact with live young poultry originating from one mail-order hatchery in the western United States, referred to as Hatchery C. Critical aspects of the investigation are ongoing surveillance, which enabled the detection of the outbreak of human disease; investigation of the illnesses, including interviews with patients, a 006 survey of agricultural feed stores, a trace-back investigation of the implicated live young poultry to identify the mail-order hatchery source, and multiple on-site inspections at Hatchery C; and ongoing interventions at Hatchery C to reduce the risk of human salmonella infections. Me thods Detection of the Outbreak In April 005, the Colorado Department of Public Health and Environment identified three Salmonella enterica serotype Montevideo isolates with identical patterns on pulsed-field gel electrophoresis (PFGE) (XbaI pattern JIXX0.0049, the outbreak strain). State and local public health agencies in Colorado interviewed the patients; all three reported exposure to chicks or ducklings purchased from agricultural feed stores in the week preceding the onset of illness. Queries to PulseNet, the national molecular subtyping network for foodborne disease surveillance, indicated that in 005 the outbreak strain had been isolated from five other persons in four states (Kansas, New York, Oklahoma, and Texas). To gain historical perspective on the occurrence of the outbreak strain, we queried the PulseNet database for isolates that had been reported between 999 and 004. Among the 90 salmonella Montevideo isolates reported during this period, only 5 (0.3%) were identified as 066 n engl j med 366; nejm.org may 3, 0 Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

3 Salmonellosis Outbreak Linked to Poultry Hatchery the outbreak strain, making this a rare PFGE pattern. Surveillance and Epidemiologic Investigation As part of routine surveillance, salmonella isolates from ill persons were serotyped and subtyped with the use of PFGE at state and local public health laboratories, and the results were submitted to PulseNet. A case was defined by a clinical sample yielding the outbreak strain of salmonella Montevideo and uploaded to the PulseNet database between, 004, and December 3, 0. Testing for susceptibility to a standard panel of 5 antimicrobial agents was performed by the National Antimicrobial Resistance Monitoring System. State and local public health officials interviewed patients or their caretakers, using a series of standardized questionnaires developed by the Centers for Disease Control and Prevention (CDC), state and local health departments, and the NPIP. The questionnaires focused on clinical illness, contact with live poultry during the week preceding the onset of illness, details related to the purchase of poultry, the patient s or caretaker s knowledge of salmonella transmission, and the extent to which the patient or caretaker was informed at the time of purchase about the risk of acquiring salmonella from live poultry. As the information available over the 8 years of the investigation evolved, additional questions were asked of patients. Analyses of the database were performed by CDC personnel. All study participants (patients or caretakers) provided oral informed consent. Survey of Agricultural Feed Stores in New Mexico In March 006, we surveyed representatives of agricultural feed stores in New Mexico that sold young poultry during that year. New Mexico was chosen because it had a large number of cases as well as resources available to support investigation activities. Stores identified in an Internet search were randomly selected for an in-person or a telephone interview. The standardized questionnaire focused on the source of the live poultry, volume of live-poultry sales, handling and hygienic conditions of poultry in the store, knowledge about the risk of transmission of salmonella from poultry to humans, and education of customers about this risk. Trace-back Investigation Information provided by patients was used to trace the young poultry they had purchased through retail outlets to the mail-order hatchery of origin. State and local public health officials and animal health agencies conducted the traceback investigations. R esult s Characteristics of the Patients From 004 through 0, the outbreak strain was isolated from 36 case patients, as reported by the departments of health in 43 states (Table and Fig. ). The number of cases peaked annually in the spring; the greatest number of cases was reported in 006 (Fig. ). Age and sex were reported for 66 and 80 patients, respectively. The median age of patients was 4 years (range, month to 86 years), and 43 (54%) were 5 years of age or younger (Table and Fig. 3); 49 patients (53%) were female. Interviews were completed with 56 (49%) of the 36 patients (or caretakers); 53 (98%) reported diarrhea, and among the 45 patients with available information, 80 (55%) reported bloody diarrhea. Thirty-six of the 56 patients interviewed (3%) were hospitalized for a median duration of days (range, to 4). Twenty-three (64%) of the 36 hospitalized patients were 5 years of age or younger. No deaths were reported. All isolates were susceptible to the standard panel of 5 antimicrobial agents. Information on contact with live poultry was available for 59 patients, and of these patients (77%) were reported to have had contact with live young poultry during the week preceding the onset of illness (Table ). Among the 89 patients or caretakers who were asked whether the young poultry were artificially colored, (%) reported that they were; 0 of these patients (9%) were 5 years of age or younger. Investigators asked 9 patients or caretakers if they had purchased chicks as pets, and 38 (4%) said yes; among 3 patients or caretakers who were asked the age of the pet owner, (69%) said it was a child 5 years of age or younger. When interviewees were asked if they had known that salmonella infection could be acquired from contact with poultry, 6 of 76 respondents (%) said yes, and 5 respondents (7%) reported receiving an oral or written warning at the time of purchase on the risk of acquiring salmonella infection from live poultry. n engl j med 366; nejm.org may 3, Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

4 Table. Selected Demographic and Clinical Characteristics of Patients in the Outbreak,, 004, to December 3, 0.* (N = 36) Variable No. of states reporting to PulseNet No. of patients with available data on age Age Median yr (N = 9) 00 (N = 57) 009 (N = 5) 008 (N = ) 007 (N = 64) 006 (N = 84) 005 (N = 53) 004 (N = ) Range 80 yr mo 8 yr mo 86 yr 3 mo 85 yr mo 55 yr mo 86 yr mo 75 yr mo 77 yr mo 86 yr Patients 5 yr of age no./total no. (%) 0/ /37 (57) 37/65 (57) 9/56 (5) 5/8 (63) 6/5 (40) 7/55 (49) 8/8 (64) 43/66 (54) Female patients no./total no. (%) 0/ /4 (54) 45/7 (6) 9/6 (48) 4/6 (67) 8/5 (53) 8/55 (5) 3/8 (46) 49/80 (53) Patients interviewed no./total no. (%) 0/ 8/53 (34) 48/84 (57) 38/64 (59) 8/ (67) 5/5 (33) 8/57 (49) /9 (38) 56/36 (49) Patients with diarrhea no./total no. (%) NA 8/8 (00) 48/48 (00) 38/38 (00) 7/8 (88) 4/4 (00) 6/7 (96) /3 (9) 53/56 (98) Patients with bloody diarrhea no./total no. (%) NA 0/8 (56) 5/48 (5) 5/8 (54) 6/8 (75) /5 (40) 6/7 (59) 6/ (55) 80/45 (55) Hospitalized patients no./total no. (%) NA 6/8 (33) 8/48 (7) 8/38 () 4/8 (50) /5 (0) 7/8 (5) / (8) 36/56 (3) 5 yr of age NA 5/6 (83) 5/8 (6) /8 (5) 4/4 (00) 0/ 5/7 (7) / (00) 3/36 (64) * Unless noted otherwise, the information provided was obtained from the PulseNet database. NA indicates that data were not available. Information was obtained from patient interviews conducted by local and state departments of health. Survey of Agricultural Feed Stores in New Mexico In 006, we surveyed representatives from 54 agricultural feed stores identified as selling chicks in New Mexico; half were surveyed in the store and half by telephone. The median number of chicks sold per store was 600 per year (range, 50 to 5500). Of the 54 respondents, (39%) reported that chicks from different shipments were comingled at the store; 3 (57%) reported that the bird housing area was cleaned daily, (0%) reported that it was disinfected between shipments, and 9 (35%) reported that it was disinfected only at the end of the season. Artificially colored chicks were sold by 3 of the stores (59%). Representatives from 46 of the stores (85%) were aware that live young poultry can transmit salmonella to humans; among these representatives, 6 (57%) reported that they give customers oral or written warnings about transmission. Trace-Back Investigation Between 005 and 0, trace-back investigations were conducted for live poultry found in the households of 59 case patients to determine the hatchery origin. In 48 of these households (8%), the origin was Hatchery C. Seven other mail-order hatcheries were also identified during the investigations. Hatchery C is known to drop-ship birds for other mail-order hatcheries; however, investigators were unable to determine whether implicated birds were drop-shipped from Hatchery C. Hatchery Investigation and Interventions Hatchery C ships birds to customers in all 50 states and produces an estimated 4 million birds annually. During the spring, which is the peak shipping season, Hatchery C ships an estimated 50,000 birds in one week (unpublished data). In 006, poultry experts and veterinary consultants began providing technical assistance to the owners of Hatchery C for the prevention and control of salmonella. Specific recommendations were provided for increased biosecurity, enhanced rodent control, feed decontamination, routine surveillance of the hatchery environment through microbiologic sampling at the premises, and the use of autogenous poultry vaccine. The owners of Hatchery C voluntarily implemented recommendations by replacing, updating, and sealing old equipment and floors, changing airflow within the facility, implementing a quater- 068 n engl j med 366; nejm.org may 3, 0 Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

5 Salmonellosis Outbreak Linked to Poultry Hatchery Figure. Location and Number of Patients in the Outbreak According to State,, 004, to December 3, 0. A total of 36 cases were identified. The data are shown according to the state that tested the patient for the isolate strain and uploaded the data to PulseNet. nary ammonium egg-cleaning procedure, improving hatchery biosecurity, conducting routine biologic surveillance for salmonella on the hatchery premises, and contracting with a private company to develop an autogenous poultry vaccine designed to reduce the presence of the outbreak strain in the hatchery. Widespread vaccination of all adult chickens at Hatchery C began in August 006. In October 007, an investigation team consisting of representatives from the NPIP, state departments of public health and animal health, and the CDC visited Hatchery C to monitor the cleanup progress, make further recommendations on salmonella prevention and control, and collect additional biologic samples from birds, droppings, and litter and in areas or on equipment with which birds had contact in order to test for the presence of bacteria in the environment. Collected samples were cultured for salmonella at three independent laboratories. Serotyping and molecular subtyping with the use of PFGE were performed at the National Veterinary Services Laboratories of the USDA Animal and Plant Health Inspection Service. The outbreak strain was identified in 58 isolates from these samples by PulseNet. Between 008 and 009, an additional 00 environmental samples were collected from the premises of the hatchery, cultured for salmonella at a commercial laboratory, and serotyped and subtyped with the use of PFGE at the state public health laboratory. The outbreak strain was detected in 4 (7%) of these samples. A veterinary consultant continued to visit the hatchery, monitoring surveillance findings and suggesting additional interventions as necessary. Discussion In this report, we describe a prolonged and ongoing multistate outbreak of human salmonella Montevideo infections primarily affecting young children and linked to contact with live young poultry from a single mail-order hatchery. Trace- n engl j med 366; nejm.org may 3, Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

6 No. of Isolates Uploaded Outbreak detected 005 Intervention begins at Hatchery C Figure. Number of Patients Infected with the Salmonella Outbreak Strain, According to Date of Report to PulseNet Database,, 004, to December 3, 0. No. of Cases Age Group (yr) Figure 3. Number of Patients in the Outbreak According to Age Group,, 004, to December 3, 0. Information on age was available for 66 of the 36 patients identified in the PulseNet database. back investigations of poultry to which ill persons were exposed identified Hatchery C, a mailorder hatchery in the western United States, as the source of the birds. During the 8-year period from 004 through 0, we identified 36 cases in 43 states as part of the outbreak. Because only a portion of salmonella infections are laboratoryconfirmed, it is likely that thousands of additional unreported infections occurred in association with this outbreak. Although the number of reported human infections with the outbreak strain declined after the implementation of enhanced biosecurity and intervention measures at the hatchery in 006, cases continued to be reported. It is likely that a temporary lapse in the hatchery s standard operating procedures in 00 resulted in increased bacterial loads in the hatchery environment. Findings from epidemiologic and environmental investigations and the fact that Hatchery C has a closed breeding facility that does not introduce birds from outside flocks suggest that this outbreak strain persisted in the hatchery s environment throughout the 8-year investigation. At the time of this report, Hatchery C was working closely with the state departments of public health and animal health, a veterinary consultant with poultry expertise, the CDC, and the NPIP to enhance and further implement salmonella prevention and control recommendations, with the goal of preventing additional human infections. Hatchery C is not the only hatchery in the United States that has been linked to outbreaks of human salmonellosis caused by infected poultry. Reports of investigations conducted from 955 through 009 describe outbreaks of human salmonellosis involving multiple salmonella serotypes and mail-order hatcheries in the East, the Midwest, the Southwest, and the Pacific Northwest. -8 The current investigation shows that control interventions at mail-order hatcheries can play a role in the reduction of salmonella transmission and suggests that consumer education and hygienic interventions at venues selling young poultry are measures that have not been fully incorporated into a comprehensive prevention strategy. The investigation also shows the difficulty of reducing salmonella transmission from live poultry. Between and April 30, 0, only one human case of infection with the outbreak strain was reported to PulseNet. The trend of decreasing numbers of cases reported after the interventions at the mail-order hatchery and the low number reported in the first 4 months of 0 support the conclusion that the interventions are helping to reduce human infections. Consumers wishing to reduce their risk of illness should practice meticulous hand hygiene 070 n engl j med 366; nejm.org may 3, 0 Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

7 Salmonellosis Outbreak Linked to Poultry Hatchery Table. Selected Characteristics of Patient Exposure in the Outbreak,, 004, to December 3, 0.* Variable Patients reporting contact with live young poultry no./total no. (%) Type of poultry no./total no. (%) NA 4/8 (78) 4/48 (88) 30/38 (79) 7/8 (88) 3/5 (60) 5/8 (54) /4 (79) /59 (77) Chicks NA 9/4 (64) 30/4 (7) 3/8 (8) 3/7 (43) 3/3 (00) /5 (73) 0/ (83) 89/ (74) Ducklings NA 9/4 (64) 3/4 (3) 6/8 () 6/7 (86) 0/3 5/5 (33) / (8) 40/ (33) Goslings NA /4 (7) 5/4 () 0/8 0/7 0/3 0/5 0/ 6/ (5) Contact occurred at home no./total no. (%) NA /4 (86) 6/4 (6) 7/6 (65) 4/7 (57) / (00) /5 (80) 8/ (73) 8/7 (69) Poultry kept inside living space no./total no. (%) NA 7/4 (50) 7/4 (40) 4/4 (7) /4 (50) / (00) 4/5 (7) 7/ (64) 43/ (38) Poultry purchased at agricultural feed store no./total no. (%) NA /4 (86) 3/34 (94) 6/30 (87) 6/7 (86) / (00) 0/5 (67) 3/3 (00) 9/05 (87) Contact with artificially colored poultry no./total no. (%) NA 3/4 () 8/4 (9) NA 0/7 0/ 0/4 0/0 /89 () Child 5 yr of age NA 3/3 (00) 7/8 (88) NA NA NA NA NA 0/ (9) Purpose of purchase no./total no. (%) Meat NA NA 4/4 (33) NA 0/7 0/ 0/5 3/ (7) 7/77 () Eggs NA NA 7/4 (7) NA /7 (4) / (50) 9/5 (60) 9/ (8) 7/77 (35) Pets NA 8/4 (57) 8/4 (43) NA 4/7 (57) 0/ 5/5 (33) 3/ (7) 38/9 (4) Child 5 yr of age NA NA /8 (67) NA 4/4 (00) 0/ 4/5 (80) /3 (67) /3 (69) Patients who knew salmonella can be transmitted by contact with live poultry no./total no. (%) Patients receiving warning about salmonella transmittal at time of purchase no./total no. (%) NA /3 (5) 7/48 (5) NA NA NA 7/5 (47) NA 6/76 () NA /4 (7) 3/4 (7) NA NA NA /5 (7) NA 5/7 (7) * All information was obtained from patient interviews. NA denotes not available. n engl j med 366; nejm.org may 3, 0 07 Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

8 and encourage this behavior in children. High-risk groups, including children younger than 5 years of age, elderly persons, and immunocompromised persons, should not handle or touch chicks, ducklings, or other live poultry. Live poultry should not be allowed inside a residence, in bathrooms, or in areas where food or drink is prepared, served, stored, or consumed. Any equipment or materials associated with raising or caring for live poultry should be cleaned outside the residence. Additional recommendations on reducing the risk of illness associated with live poultry are available at posters.htm. This investigation shows that the sanitary conditions of housing for birds at agricultural feed stores vary and that the comingling of birds from multiple shipments occurs. Comingling can be a source of the propagation of pathogens among birds that are sold to the public. At this time, no standard guidelines are available for owners of agricultural feed stores on the implementation of effective hygienic and biosecurity measures in areas housing birds (unpublished data). We recommend that public health and animal health officials and industry partners develop guidelines to reduce salmonella transmission. Approximately half the ill persons in this outbreak were children 5 years of age or younger. We recommend prohibiting practices (e.g., artificially coloring birds) that target the sale of birds to children. This investigation shows that consumers are often unaware of the risk of salmonellosis from exposure to live poultry and are not adequately warned at the time of purchase. Staff at agricultural feed stores should encourage customers to wash their hands after touching live poultry or objects in bird-display areas, and information warning customers about the risk of salmonella infections from live poultry should be distributed at venues where live poultry are sold. 3 The CDC, the NPIP, and some departments of health have developed educational material on the risk of illness associated with live poultry. 4,5 Comprehensive control programs should be developed and maintained at hatcheries and breeder farms. These programs should include routine microbiologic monitoring and effective and practical sanitation components directed at reducing the occurrence of salmonella organisms in hatching eggs and chicks. The NPIP has established sanitation standards for poultry breeding stock and hatchery products to prevent the spread of diseases that are transmitted by eggs and disseminated at hatcheries. 6 Mail-order hatcheries may voluntarily choose to follow NPIP recommendations for meeting these standards; however, wide variation among hatchery operations makes standardization of hatchery cleanup protocols difficult, and the high cost of implementation makes it difficult for small hatcheries to comply. We encourage the NPIP, industry partners such as mail-order hatchery owners, and public health and animal health agencies at the local, state, and federal levels to continue to collaborate in developing and implementing comprehensive prevention, control, and educational measures from the hatchery to the consumer in order to reduce the risk of human salmonella infections associated with live poultry. Information warning customers about the risk of salmonella infections from live poultry should be distributed by mailorder hatcheries with all shipments of live birds. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. Presented in part at the Council of State and Territorial Epidemiologists Annual Conference, Atlantic City, NJ, June 4 8, 007; at the American Veterinary Medical Association Convention, St. Louis, July 6 9, 0; and at the U.S. Animal Health Association Conferences, Reno, NV, October 8 3, 007, and San Diego, CA, October 7 4, 009. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. We thank Laura Nathan (Arizona Department of Health Services); Akiko Kimura, Janet Mohle-Boetani, and Jennifer Schneider (California Department of Health Services); William Ray (Colorado Department of Public Health and Environment); Rita Bagby (County of Los Angeles Department of Health Services); Jeffrey Higa, Annie Kao, and Carol Roach (County of San Diego Health and Human Services Agency); Pat Piercy (Illinois Department of Public Health); Sara Brend (Iowa Department of Public Health); Carol Giliam and Sandy Kelly (Kentucky Department for Public Health); Emily Harvey (Massachusetts Department of Public Health); JoAnn Rudruff (Missouri Department of Health and Senior Services); Dennis Leschinsky (Nebraska Health and Human Services System); Michelle Malavet (New Jersey Department of Health and Senior Services); Jessica Jungk, Mack Sewell, and Aaron Wendelboe (New Mexico Department of Health); Hillary Booth and Julie Hatch (Oregon Public Health Division); Linda Schaefer (South Dakota Department of Health); John Dunn, Samir Hanna, and Amanda Ingram (Tennessee Department of Health); Leslie Bullion (Texas Department of State Health Services); Marilee Poulson and Diane Raccasi (Utah Department of Health); Seth Levine (Virginia Department of Health); John Archer (Wisconsin Department of Health and Family Services); Kelley Weidenbach (Wyoming Department of Health); Heather Bair-Brake, Matt Biggerstaff, Elizabeth Cavallaro, Olga Henao, Sofiya Kahn, Anagha Loharikar, Jennifer Mitchell, Thai-An Nguyen, Katherine O Connor, Colin Schwen sohn, Umid Sharapov, and Ian Williams (Division of Foodborne, Waterborne, and Environmental Diseases, Centers for Disease Control and Preven- 07 n engl j med 366; nejm.org may 3, 0 Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.

9 Salmonellosis Outbreak Linked to Poultry Hatchery tion, PulseNet); Jana Austin and Christopher Perry (Enteric Diseases Laboratory Branch, Centers for Disease Control and Prevention); Charles S. Roney (Department of Agriculture, Animal and Plant Health Inspection Service); Brett Hopkins (Pfizer); Matthew Erdman (Department of Agriculture, Animal and Plant Health Inspection Services, Veterinary Services, National Veterinary Services Laboratory); the owners and staff of Hatchery C; departments of health in the following states: Connecticut, Georgia, Indiana, Kansas, Maine, Maryland, Minnesota, New Hampshire, New York, North Dakota, Oklahoma, Pennsylvania, Rhode Island, Vermont, Washington, and West Virginia; Yvonne Butler, Kenneth Grant, and Glenn Hansen for assistance in data collection, entry, and analysis; and Henri Schumacher for editorial assistance with an earlier version of the manuscript. References. Scallan E, Hoekstra RM, Angulo FJ, et al. Foodborne illness acquired in the United States major pathogens. Emerg Infect Dis 0;7:7-5.. Anderson AS, Bauer H, Nelson CB. Salmonellosis due to Salmonella typhimurium with Easter chicks as likely source. JAMA 955;58: Salmonella hadar associated with pet ducklings Connecticut, Maryland, and Pennsylvania, 99. MMWR Morb Mortal Wkly Rep 99;4: Salmonella serotype Montevideo infections associated with chicks Idaho, Washington, and Oregon, spring 995 and 996. MMWR Morb Mortal Wkly Rep 997;46: Salmonellosis associated with chicks and ducklings Michigan and Missouri, spring 999. MMWR Morb Mortal Wkly Rep 000;49: Wilkins MJ, Bidol SA, Boulton ML, Stobierski MG, Massey JP, Robinson- Dunn B. Human salmonellosis associated with young poultry from a contaminated hatchery in Michigan and the resulting public health interventions, 999 and 000. Epidemiol Infect 00;9: Three outbreaks of salmonellosis associated with baby poultry from three hatcheries United States, 006. MMWR Morb Mortal Wkly Rep 007;56: Multistate outbreak of salmonella infections associated with live poultry United States, 007. MMWR Morb Mortal Wkly Rep 009;58: Roy P, Dhillon AS, Shivaprasad HL, Schaberg DM, Bandli D, Johnson S. Pathogenicity of different serogroups of avian salmonellae in specific-pathogenfree chickens. Avian Dis 00;45: Neuman W. Keeping their eggs in their backyard nests. New York Times. August 4, Swaminathan B, Barrett T, Hunter S, Tauxe R. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg Infect Dis 00;7: Human isolates: final report, 007. Atlanta: National Antimicrobial Resistance Monitoring System, Centers for Disease Control and Prevention, Compendium on handwashing materials. Atlanta: Centers for Disease Control and Prevention ( 4. Risk of human salmonella infections from live baby poultry. Atlanta: Centers for Disease Control and Prevention ( 5. North Dakota Department of Health and State Board of Animal Health warn residents about risk of salmonella infection from baby birds. Press release of the North Dakota Department of Health, March 0, 008 ( Salmonella/SalmonellaNewsRelease.pdf). 6. National Poultry Improvement Plan: seventy years of poultry improvement. Program aid no Washington, DC: Department of Agriculture, 006. Copyright 0 Massachusetts Medical Society. n engl j med 366; nejm.org may 3, Downloaded from nejm.org on April, 08. For personal use only. No other uses without permission. Copyright 0 Massachusetts Medical Society. All rights reserved.