Disease Control Bulletin: December 2001


disease control bulletin

Foodborne Disease Surveillance

Foodborne disease is a common problem in the United States, causing an estimated 76 million illnesses, 325,000 hospitalizations, and 5,000 deaths annually.1 Gastrointestinal symptoms such as vomiting, diarrhea, and abdominal pain are the most common presenting symptoms, though some agents cause neurologic or nonspecific symptoms. Nationally the most common pathogens associated with foodborne illness are Norwalk-like viruses; these illnesses are estimated to account for approximately two-thirds of foodborne illnesses.2 Because viral testing is not commonly done for foodborne illness, most available data relate to bacterial pathogens.

In Vermont, campylobacteriosis, salmonellosis, E. coli O157:H7 infections, yersiniosis, and shigellosis are the most commonly reported foodborne diseases (see below). While these infections can be transmitted by other routes, such as person-to-person contact or handling animals, foodborne transmission is the primary risk. Less common foodborne illnesses reported between 1996 and 2000 in Vermont have included listeriosis (12 cases), Vibrio vulnificus infection and typhoid fever (Salmonella typhi), two cases each; and one case of trichinellosis.

Foodborne Diseases in Vermont 1996-2000
Agent Number of cases/yr (5-year median) Range
Campylobacteriosis 143 102–173
Salmonellosis 101 88 – 144
E. coli O157:H7 32 8 – 37
Yersiniosis 13 4 – 15
Shigellosis 7 4 – 12

In addition to specific diseases, any unexpected pattern of cases or suspected cases of illness, which might indicate an outbreak, are reportable under Vermont’s Communicable Disease regulations. Between 1996 and 2000, there were 21 foodborne outbreaks investigated in Vermont, with between one and seven outbreaks investigated per year. Ten outbreaks of acute gastrointestinal illness of unknown etiology occurred, most probably due to viral agents. Other agents involved in outbreaks included Salmonella (six outbreaks), cyclospora (two) hepatitis A (one), E. coli O157:H7 (one), and Campylobacter (one).

Physicians and other health care providers play a critical role in surveillance for foodborne disease problems. Considering the possibility of foodborne disease and reporting suspicious cases or clusters of cases to the Health Department allows for the immediate investigation of a possible foodborne disease outbreak. Stool tests might confirm a specific bacterial or parasitic etiologic agent. For the individual patient, identification of the specific agent causing illness is important for the appropriate management of the illness. For public health, identifying the causative agents of individual illnesses provides data for identifying disease trends and disease burden.

Surveillance for foodborne disease is also the basis for recognizing outbreaks. Prompt investigation of outbreaks can lead to the identification and elimination of harmful food items or risky food handling practices. Immediate benefits can include temporary removal of ill persons from high-risk occupations, provision of treatment of contacts when appropriate (e.g., immunoglobulin for prevention of hepatitis A), and reinforcement of measures such as handwashing to decrease transmission to household contacts.

CDC Primer on Foodborne Illnesses

In January 2001, the CDC published “Diagnosis and Management of Foodborne Illnesses: A Primer for Physicians.”3 This document is currently available on the Ameri-can Medical Association’s website at www.ama-assn.org/ama/ pub/category/3707.html. The primer includes a section on general clinical considerations, followed by foodborne illness tables for bacterial, viral, parasitic, and non-infectious agents. For each agent the tables provide information on incubation periods, signs and symptoms, duration of illness, frequently associated foods, laboratory testing, and treatment options.

The primer recommends stool cultures if a patient is febrile, has bloody diarrhea, has severe abdominal pain, has many fecal leukocytes in the stool, is immunocompromised, or if the illness is clinically severe or persistent. Public health benefits from health care providers having a low threshold for requesting stool cultures. Any individual patient seen could be the index case for a foodborne disease outbreak, since documented cases represent only a small portion of foodborne illness. In the 1998 Vermont Behavioral Risk Factor Surveillance Survey randomly selected adults were asked if they had experienced diarrhea in the past 30 days; 13.4 percent (n=407 ) said that they had. However among those people only 66 (14.6%) reported contacting their health care provider about the diarrhea, 25 were asked to submit a stool sample and 22 actually submitted a test. Thus, testing was only done on 5.4 percent of those who were ill.

The classic foodborne disease outbreak stems from a contaminated food item served at a single event. This “point source” outbreak often leads to an easily identified cluster of illnesses. With current food distribution practices, items grown or processed at one location are frequently widely distributed, both nationally and internationally. Low-level contamination of such a product can cause sporadic-appearing illnesses. It is much more difficult to determine the source of these infections. However, advances in laboratory testing, including new molecular techniques, are enabling epidemi-ologists to link cases caused by a common source.

Clinical laboratories are strongly encouraged to send all Campylobacter, Salmonella, Shigella, E. coli O157:H7 and Listeria isolates to the Vermont Department of Health Laboratory for reference testing. The state Laboratory provides serotyping for reference cultures of enteric bacterial agents and molecular subtyping for Salmonella and E.coli O157:H7 (described below). Salmonella serotyping alone does not usually provide sufficient differentiation for epidemiologic purposes. The majority of Salmonella infections, both nationally and in Vermont, are caused by the two most common serotypes, Salmonella typhimurium and Salmonella enteriti-dis. In Vermont, for the year 2000, S. typhimurium accounted for 49 (44.5%) Salmonella isolates, followed by S. enteriti-dis with 20 (18.1%) isolates. Most of these cases did not demonstrate an epidemiologic connection.

PFGE Testing

The ability to link foodborne bacterial infections has been greatly enhanced through the use of a molecular technology called pulsed-field gel electrophoresis (PFGE) testing. The Vermont Department of Health began PFGE testing in April 2000 and currently performs PFGE on all nontyphoidal Salmonella serotypes and E. coli 0157:H7 isolates received at the Health Department Laboratory. PFGE was performed on all isolates for both Salmonella and E coli O157:H7 received since July 1, 1999 to create a reference database containing all Vermont patterns seen during that time period. Each new pattern is compared to the database and given a pattern designation depending upon whether or not it matches an existing pattern in the database. All matching patterns are given the same pattern designation, allowing the microbiologist to detect clusters or outbreaks more readily.

In PFGE testing, bacterial DNA is first cut into small pieces using a restriction enzyme. The pieces are then separated on a gel. The DNA is placed at one end of the gel and an electric field is applied across the gel, driving the pieces of DNA from one end to the other. The smaller fragments of DNA migrate faster through the gel than the larger fragments. The pieces are separated as distinct bands on the gel. The resulting pattern is the “fingerprint.” Once the patterns have been generated they are transmitted to CDC where they are filed in a central PulseNet database.

PulseNet is the national molecular subtyping network for foodborne disease surveillance. It was developed by CDC and several state health department laboratories to facilitate subtyping bacterial foodborne pathogens for epidemiologic purposes. PulseNet participants have the ability to exchange information electronically, so laboratorians and epidemiolo-gists in different states can compare DNA fingerprints to those in the PulseNet database, leading to enhanced outbreak detection. Each participating laboratory strictly follows the standardized protocols developed by CDC. The results from every lab are highly reproducible and the DNA patterns produced at each lab can be compared at other participating PulseNet laboratories and at CDC. The use of standardized subtyping methods allows isolates from different parts of the country to be compared, enabling recognition of nationwide outbreaks. Bacterial strains from suspected food items can be compared to strains from humans. Matching patterns could indicate a nationwide outbreak and lead to epidemiologic investigations and product recalls.

PFGE patterns are useful in providing an early focus for an investigation. For example, during the summer of 2000, the Health Department received reports of three cases of E. coli O157:H7 in a two-day period. Two cases were children living in the same town; the third case was another child attending a camp in a neighboring town. It is very unusual to have three cases of this infection reported in such a short time frame. The patients were quickly interviewed in order to determine a possible common source of infection; no obvious connections were identified. However, PFGE results were soon available, and indicated that despite the clustering in time, these cases had distinct patterns. This allowed Epidemiology to separately investigate the camp. Other ill children were identified, and the subsequent outbreak investigation implicated ground meat as the probable source of infection.


  1. United States General Accounting Office. Report on food safety. GAO-010973. Washington, D.C. September 2001
  2. Mead PS, Slutsker L, Dietz V, et al. Food-related illness and death in the United States. Emerg Infect Dis 1999;5:607-17.
  3. Centers for Disease Control and Prevention. Diagnosis and Management of Foodborne Illnesses: a Primer for Physicians. MMWR 2001;50(No.RR-2)


Pertussis: Not Just for Kids

Between Novemter 13 and Deeember 6, 2001, at least 11 confirmed cases of pertussis were identified in Vermont students. Many more students are being evaluated, treated for cough illness, or prophylaxed as contacts.

From January, 1998, through September, 2001, 457 confirmed cases of pertussis were reported in Vermont. Twenty-nine of these cases were infants less than 1 year old (eight in 1998, seven in 1999, 10 in 2000, and four so far in 2001). Illness in this age group is associated with the potential for significant complications, such as prolonged apnea, and an increased risk of death. Fourteen (48%) of the 29 infants required hospitalization, with hospital stays ranging from two to 50 days. Most (26, or 90%) of the infants were less than 6-months-old (three were <1-month-old, 11 were 1-month-old, two were 2-months-old, three were 3-months-old, one was 4-months-old, and six were 5-months-old). These infants would not have been fully immunized at the time of their illnesses even if their immunizations were up-to-date.

Why does pertussis continue to occur in spite of high immunization levels identified among Vermont 2-year-olds? Adolescents and adults are also susceptible to pertussis and can serve as the unrecognized source of infection for infants and children who are unimmunized or incompletely immunized. Several studies have reported isolation of Bordetella pertussis from adults with a persistent cough, i.e., a cough that has lasted 2 or more weeks. In college students, for example, at least 25 percent of persistent cough illness is likely due to pertussis. During interviews with the contacts of 18 of the infants in Vermont mentioned above, investigators established that 14 (78%) of the 18 cases had had exposure to an individual with a prolonged cough; half of these contacts were adults.

Prevention involves several strategies

The major preventive intervention is immunization. Whole-cell pertussis vaccine was developed in the 1940s, and acellular pertussis vaccine, which has fewer associated adverse effects, was licensed in 1991. Nationally, the number of pertussis cases has decreased from pre-vaccine levels of greater than 200,000 cases per year to a low of just over 1,000 cases in 1976. Since the mid-1990s, however, the number of cases has been increasing.

Administration of pertussis vaccine is recommended for children at ages 2 months, 4 months, 6 months, 15–18 months, and 4–6 years. Immunity to pertussis as elicited by vaccination is not lifelong, however. Booster vaccination is not currently recommended for persons over age 6 years because vaccine reactions are possibly more frequent in older individuals and the morbidity and mortality from pertussis decrease with increasing age. The overall decreased incidence of pertussis in the general population limits the re-exposure

to disease that was experienced in the past by older children and adults and acted as a booster to waning immunity. Recent outbreaks of pertussis in teens highlight the susceptibility of these individuals to this disease. Studies are currently underway to determine if the acellular pertussis vaccine will be safe and effective in older children and adults.

Another preventive intervention is chemoprophylaxis of susceptible individuals who have been exposed to a person with pertussis. This is routinely undertaken when a child is diagnosed with pertussis; the household, day care, and/or close school contacts are identified and prophylaxed according to recommendations from the Advisory Committee on Immunization Practices from the Centers for Disease Control and Prevention. Too often, however, pertussis in an adult is undiagnosed, and contacts are neither identified nor treated.

Diagnosing pertussis—yes, even in adults

Most important for diagnosing pertussis is the index of suspicion that a cough illness might be pertussis, in adolescents and adults as well as in children. Classic presentation of the illness in children includes paroxysmal cough (spasms of coughing), the characteristic “whoop” following cough, and post-tussive vomiting. In adults, the main presentation might be persistent cough. Providers need to have a lowered threshold for testing for pertussis in adults if the patient has contact, either at work or at home, with infants, especially infants less than 6 months of age.

Whenever pertussis is considered in the differential diagnosis of a cough illness, providers need to confirm the diagnosis with culture and/or PCR testing of a properly taken nasopharyngeal specimen. Specimens for culture must be taken before the individual is on antibiotics. B. pertussis can be slow-growing, and a culture must be held for at least seven and up to 14 days before being read as negative. Pertussis cultures are performed at the Vermont Department of Health Laboratory.

PCR might demonstrate the presence of the organism, even after antibiotics have been taken. The PCR test is expensive and, since it is not currently performed at any laboratories located in Vermont, specimens must be sent out of state. Results of DFA testing are no longer included in the case definition for pertussis because of limitations associated with interpreting test results. Serology is not acceptable for diagnosing pertussis (except in Massachusetts, where there is an extensive in-state program for performance and evaluation of serological testing).

Proper diagnosis and evaluation of pertussis in children, adolescents, and adults will help to protect Vermont’s babies.


MMWR 1991:40(RR10):1-28


Chronic Disease in Vermont: Asthma 2001

Asthma in Children

The Vermont estimated prevalence of pediatric asthma was 51 per 1,0001 for children less than 18 years old according to the 1998 National Health Interview Survey. However, preliminary unweighted Vermont data suggest childhood prevalence may be higher (77.3 per 1,0002). Nationally, childhood asthma prevalence of 120.6 per 1,000 in 19983 with a 95 percent confidence interval (CI)[113.8,127.5], is approximately double or triple the prevalence estimated for Vermont children. The rate of asthma hospitalization4 for Vermont children under 15 years of age decreased significantly from 1997 (1.2 per 1,000 95%CI [0.99,1.37]) to 1998 (0.77 per 1,000 95%CI [0.61,0.93])(see Figure 1). Vermont rates were approximately one-third of the U.S. hospitalization rates.3 There were no asthma deaths among Vermont children under age 15 during the time period 1994-1998.5

Asthma in Adults

The 1999-2000 average prevalence of asthma in adult Vermonters was 7.5 percent with a 95%CI [6.8,8.3]. Men had a significantly lower prevalence of asthma (6.2%, 95%CI [5.2,7.1]) than women (8.9%, 95%CI [7.8,10.0]). In 1998, the 65+ year old age group became the Vermont population with the highest hospitalization rates for asthma (1.0 per 1,000 95%CI [0.77,1.22])(see Figure 1), but the rate was significantly lower than the U.S. hospitalization rate (1.8 per 1,000)3 for this age group. In 1998, seven adult Vermonters died from asthma. The average age-adjusted6 mortality rate for the state (1994-98) was 1.2 per 100,000, lower than the 1998 U.S. rates for white race only (1.7 per 100,000) and all races combined (2.0 per 100,000).

asthma chart

Characteristics of Adult Vermonters with Asthma

Of those with asthma, 24.6 percent (95%CI [20.3,29.0]) were obese (body mass index 30 or more) compared to 18 percent of all Vermonters (see Figure 2). Most (78.1%) adult Vermonters with asthma did not meet the recommended level of physical activity,2 28.6 percent had allowed smoking in their homes in the previous 30 days, 22.4 percent were current smokers, 73 percent were in the middle socioeconomic status, 2 and 9.6 percent were lacking health insurance. There were no significant differences in these characteristics for those with asthma versus all adult Vermonters.2

asthma chart 2* Depression defined as 2 or more weeks within the past year or two years of life feeling depressed or sad AND more than one day of the past week feeling depressed.

People with asthma represent one of the target groups for vaccination because of their increased risk for complications from influenza,7 yet only 40.7 percent (95%CI [35.8,45.5]) of adults with asthma had a flu shot in the past twelve months.

People with current asthma reported fair or poor health status2 (as opposed to good or excellent health) more often (21.2%, 95%CI[17.3,25.1]) than those who did not have asthma at all (9.7%, 95%CI[8.8,10.4). People with asthma were more likely to report experiencing depression2 (24.1%, 95%CI[19.7,28.4]) than people who did not have asthma (11.6%, 95%CI[10.6,12.5]).

Asthma Advisory Panel

A Vermont Asthma Advisory Panel has been formed with funding from the Centers for Disease Control and Prevention (CDC) and is working in the following areas: Clinical Issues; Policy, Payer, and Data Issues; School-based Interventions; and Environmental Interventions. Contact Norma Wasko at (802) 865-7744 for more information.


  1. American Lung Association. The Estimated Prevalence and Incidence of Lung Disease by Lung Association Territory. April 2001.
  2. Vermont Behavioral Risk Factor Surveillance System draws from the non-institutionalized state population aged 18+ years. Pediatric data are from the first nine months of 2001. Physical Activity: five or more 30 minute sessions of exercise per week. Middle SES: more than high school and less than college graduation or income $19,999 to $34,999).
  3. American Lung Association. Trends in Asthma Morbidity and Mortality. Epidemiology & Statistics Unit. January 2001. US Hospitalization rates (under 15) were 3.6 per 1,000 in 1997 and 2.8 per 1,000 in 1998.
  4. Vermont Hospital Discharge Data Set.
  5. Vermont Vital Statistics Data Set.
  6. CDC WONDER. Age-adjusted to the US Standard 2000 population.
  7. Centers for Disease Control and Prevention. Prevention and control ofinfluenza: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 2001;50(No. RR-4):8.


75th Anniversary of Communicable Disease Control in Vermont Public Health

According to “A Chronological History of the Vermont Department of Health 1872-1957,” compiled in May 1957 by Viola Russell, MD, then Director of the Division of Maternal and Child Health, the Division of Communicable Disease Control was organized in 1926 to “pick up the Venereal Disease Program with matching federal funds for an expanding general epidemiological program.” Infectious Disease Epidemiology now resides within the Division of Health Surveillance. Our responsibilities still include surveillance, prevention, and control of venereal diseases (now known as sexually transmitted infections), such as gonorrhea and syphilis, plus others not known or diagnosable in 1926 (such as chlamydia and HIV/AIDS). We are also responsible for surveillance, prevention, and control of vaccine-preventable diseases (such as measles, pertussis, and varicella); zoonoses (such as rabies and Lyme disease); emerging infectious diseases (such as West Nile Virus); tuberculosis; hepatitis; foodborne illnesses (such as Salmonella and E. coli O157:H7); waterborne illnesses (such as giardia and Cryptosporidium); and other infectious diseases of varying incidence. A new task for us is response to agents of bioterrorism.

The staff, including epidemiologists, nurses, a physician, a veterinarian, and administrative support staff appreciate your contributions to our work. We look forward to another 75 years of working with you to protect Vermonters from infectious disease.


New Death Certificate Form for 2002

Vermont has a new death certificate form to be used for all deaths occurring after January 1, 2002. The new Vermont Death Certificate includes a question in the Certifier section regarding the contribution of tobacco to death. New forms and instructions were mailed December 17, 2001. If you did

not receive new forms, please call the Office of Vital Records (802-863-7275) at the Vermont Department of Health . Please be sure to properly dispose of (shred or return to the Office of Vital Records) any supplies of certificates you have on hand.

1-800-640-4374 OR 1-802-863-7240

Vermont Department of Health
Division of Health Surveillance P.O. Box 70 Burlington, VT 05402-0070
Agency of Human Services
Jan K. Carney, MD, MPH


Ann R. Fingar, MD, MPH
State Epidemiologist Managing Editor