Investigation Of An Outbreak

Investigation of an outbreak

Al Katz, MD, MPH Dept. of Public Health Sciences University of Hawaii

Definition of Epidemic • The occurrence of a disease or health related event clearly in excess of normal expectation – Relative to usual frequency of occurrence in a given population, in a specified geographic area, at a certain time or season – No minimal number

Definition of “case definition” • Symptom complex with time, place, and/or person restrictions In order to count cases, must define what a case is. For foodborne outbreaks: symptom complex in association with a common meal

Definition of “attack rate” Number of new cases of illness over a specified time Population at risk

• A type of incidence rate • A measurement of the proportion of population that develops illness among those at risk • In food-borne outbreaks, compare attack rates for those eating and not eating specific items

Food-borne outbreaks • If there is a food item with a large difference in attack rate, between those who did and did not eat, incriminating evidence is provided.

Attack rates of illness according to food histories in a foodborne disease outbreak: line listing Food Item

Persons who ate specified Persons who did not eat food item specified food item Total # ill #

Punch 66 Clam dip

# well Attack Total # ill # well Attack Rate # Rate

50

16

76% 92

49 43

53%

133 97

36

73% 25

2

8%

14

61% 122 77 45

Ice 36 cream

22

23

63%

Interpreting “line listing” • Largest difference in attack rate is between those eating and not eating clam dip • Evidence that clam dip is the incriminating vehicle

Steps in an outbreak investigation 1. Establish the existence of an epidemic (may include verification of diagnosis) May need to compare the current level of the disease with past levels

Steps in an outbreak investigation 2. Describe the epidemic with respect to time, place, and person

Describe the epidemic with respect to time • Epidemic curve: Plot the number of cases (y-axis) by time (x-axis)

Number of new cases

Time

Epidemic curve • With a known “point source” exposure, an incubation period may be estimated by measuring the time of exposure to peak of epidemic curve • This knowledge can be invaluable with identification of causative agent – Staphylococcal toxin: 2-4 hours – Salmonella: 12-36 hours

Epidemic curve

Number of new cases

Meal eaten

Time

Epidemic curve

Number of new cases

Meal eaten

Peak of curve

Time

Epidemic curve: estimation of incubation period

Number of new cases

Meal eaten

Peak of curve

Time

Describe the epidemic with respect to place • Geographic distribution of cases • May be by residence, worksite, recreational site, site of common meal • When plotted, gives a “spot map” i.e., John Snow’s Broad Street cholera outbreak, London, 1854

Person • Calculate and compare attack rates by age, sex, ethnicity, occupation, even gender preference.

Steps in an outbreak investigation 3. Formulate and test hypotheses A. Is it a “common source” or “propagated” epidemic? B. Compare well and ill to see if any factors appear to be related to increased risk C. Does it appear to be related to ingesting a particular food? D. Laboratory tests: cultures, serologies, may be helpful

Steps in an outbreak investigation 4. Search for additional cases A. Alert public health authorities/clinicians so that cases may be recognized B. Look for subclinical cases among those exposed (i.e., obtain stool cultures among food handlers; serologic testing of exposed)

Steps in an outbreak investigation 5. Analysis of the data/interpretation of the findings

Steps in an outbreak investigation 5. Analysis of the data/interpretation of the findings 6. Intervention: to stop the current outbreak and/or prevent future outbreaks

Steps in an outbreak investigation 5. Analysis of the data/interpretation of the findings 6. Intervention: to stop the current outbreak and/or prevent future outbreaks 7. Final report: A. Discuss factors leading to the outbreak B. Evaluation of control measures C. Recommendations for future prevention

Honolulu Marathon foodborne outbreak, December 1987 • Saturday night, 12 December: Call to Hawaii State Department of Health from private Waikiki Clinic physician: Approximately 40 patients seen during the day with symptoms including fever, abdominal cramps, diarrhea, nausea, and vomiting. All had attended the pre-race dinner the night before

Establish the existence of an epidemic • This clearly was an increase in cases of gastroenteritis in excess of normal expectancy

Case-definition • In order to count cases, the investigators needed to create a “case definition” • Case defintion = Symptom complex with time, place, and/or person restrictions:

Symptoms of 28 runners from Japan who attended pre-race party Symptoms

Number (percent)

Diarrhea Stomach cramps Fever Headache Chills Nausea Joint pain Vomiting Muscle ache

27 (96.4) 18 (64.3) 12 (42.9) 10 (35.7) 9 (32.1) 8 (28.6) 4 (14.3) 3 (10.7) 1 (3.6)

Case definition for Honolulu Marathon pre-race party outbreak • Persons with either diarrhea, vomiting, or stomach pain who attended the pre-race party Friday, 11 December 2007

Describe the epidemic with respect to time • Epidemic curve

Epidemic curve description • Common source epidemic curve • 18 hour incubation period

Describe the epidemic with respect to place • All cases attended the pre-race party at the “Aloha Tower”

Describe the epidemic with respect to person • All persons involved attended the pre-race party • No differences in attack rate by sex or age

Formulate and test hypotheses • Common source epidemic curve with 18 hour incubation period and gastroentestinal symptoms • All persons involved had shared meal • Need to obtain food-specific attack rates for items at shared meal

Food-specific attack rates from 27 Finnish runners

Formulate and test hypotheses • Food was cultured • Stool cultures obtained from symptomatic cases and all food handlers

Search for additional cases • All Hospital Emergency Departments were notified • Public was notified by press release • Physicians were instructed on food-borne investigation techniques and asked to obtain stool cultures from cases

Data analysis • Over 280 cases were identified • Food-specific attack rates in Japanese runners pointed to salad and garlic bread • Lettuce sample culture positive for Salmonella • Stool cultures from several cases and one food handler also positive for same species of Salmonella

Intervention • Ensure all uneaten food was disposed • As outbreak had already peaked, main focus was highlighting factors related to current outbreak in order to prevent future outbreaks

Problems uncovered in the investigation • Extensive use of volunteers without adequate food handling training • Food items prepared in unlicensed setting

Final Report • Recommendations for tighter regulation of temporary food services including: • Enforcement of permit regulations: Foods must be prepared in approved kitchens • Request for increased funding to allow health department inspections for all large temporary food service operations • Proper food handling techniques must be stressed to all temporary food services permitees