Early Identification of Salmonella Cases Using Syndromic Surveillance

1
Early Identification of Salmonella Cases Using
Syndromic Surveillance Heather Brown, MPH1 , Al
Romanosky, MD, PhD1, Sadia Aslam, MPH1, Alvina
Chu, MHS2, David Blythe, MD, MPH2 Office of
Preparedness and Response1, Epidemiology and
Disease Control Program2, Maryland Department of
Health and Mental Hygiene
RESULTS Using the case finding query, 5 chief
complaints of Salmonella were found in ESSENCE
from April -August 2008. Confirmatory PFGE
testing by the State Laboratories Administration
revealed that one of these cases was associated
with the nationwide Salmonella Saintpaul
outbreak. The case visited the ED on June 6 with
a chief complaint of Salmonella in stool. The
retrospective query for Salmonella identified a
total of 3 additional chief complaints of
Salmonella since January 2007. Further
investigation of these potential cases revealed
that none were laboratory confirmed nor
associated with either outbreak.
METHODS The Maryland Department of Health and
Mental Hygiene (DHMH) conducts enhanced
surveillance using the Electronic Surveillance
System for the Early Notification of
Community-based Epidemics (ESSENCE). In Maryland,
30 of acute care hospitals in the state,
residing in the Baltimore Metro and National
Capital Region, send Emergency Department (ED)
data to ESSENCE. Although syndromic surveillance
is not generally used to detect individual cases
of disease, after the FDA alert regarding the
nationwide Salmonella outbreak, DHMH used ESSENCE
as one tool for enhanced case finding.
Previously, during the multistate outbreak of
Salmonella Tennessee associated with peanut
butter in February 2007, ESSENCE detected an
increase in persons reporting to Maryland
hospitals with chief complaints of both
Salmonella and peanut butter consumption.
Investigation of these potential cases in ESSENCE
during the 2007 outbreak was not implemented,
therefore it was not determined if these
individuals were laboratory confirmed cases of
Salmonella.

• OBJECTIVE
• The purpose of this study is to describe the use
  of syndromic surveillance emergency department
  (ED) data as a tool for enhanced case finding of
  outbreak-related illnesses.
• BACKGROUND
• On June 3, 2008, the Centers for Disease Control
  and Prevention (CDC), Food and Drug
  Administration (FDA), and public health partners
  issued a public health advisory recommending that
  consumers in New Mexico and Texas avoid eating
  certain types of tomatoes. On June 7, the warning
  was expanded and consumers were alerted of a
  nationwide outbreak of Salmonella Saintpaul
  infections.
• On June 13, Maryland confirmed its first case of
  Salmonella Saintpaul infection matching the
  outbreak strain. Based on several multistate
  case-control studies, as well as environmental
  and traceback investigations, a nationwide
  jalapeño pepper advisory was issued on July 9.
  The outbreak investigation by CDC and its public
  health partners on the federal, state, local, and
  tribal levels showed that jalapeño peppers were a
  major source of contamination and that serrano
  peppers also were a source. In addition, tomatoes
  were possibly a source, particularly early in the
  outbreak.
• The last update from CDC on August 25, 2008
  reported that the outbreak appeared to be over.
  Since April, 1442 persons infected with
  Salmonella Saintpaul with the same genetic
  fingerprint had been identified in 43 states, the
  District of Columbia, and Canada. During this
  nationwide outbreak, Maryland identified a total
  of 39 cases of Salmonella Saintpaul infections
  through traditional surveillance.

Figure 3 Tomato and Salmonella Chief Complaint
Query in ESSENCE, April August 2008
Figure 2 Peanut Butter Chief Complaint Query in
ESSENCE, February 2007
CONCLUSION ESSENCE is designed to identify
clusters of disease, but in this study ESSENCE
was tested as a tool to enhance case finding in a
national outbreak investigation. One concern
about the use of ED chief complaints for
situational awareness is that in the midst of a
well publicized outbreak, many unaffected people
may present to the ED with a chief complaint
containing a term, like Salmonella that could
mischaracterize the situation. Despite
widespread publicity about the outbreak, ESSENCE
did not detect large numbers of chief complaints
related to Salmonella. Syndromic surveillance did
detect one Salmonella case related to the 2008
nationwide outbreak among the 5 chief complaints
that contained Salmonella. If the overall
number of such chief complaints is small,
interviewing the relevant case-patients based on
their chief complaint alone may be reasonable.
In this situation only one outbreak-associated
case was identified, therefore the success for
identification of actual cases was low. However,
this Salmonella case was detected in ESSENCE one
week earlier than its report date to the state
health department. Interviewing case-patients
about possible exposures and food consumption
histories as quickly as possible is crucial to a
successful foodborne outbreak investigation.
Reducing the amount of time from infection to
recognition may minimize recall bias, improve
clinical and exposure histories, and allow public
health to identify possible sources of illness
and intervene as soon as possible to prevent
ongoing exposure. The ability to search for
specific chief complaints of illness in ESSENCE
makes the system a useful tool in enhanced case
finding during special situations, especially
those situations in which reducing the time from
infection to recognition of cases is critical for
outbreak investigation and mitigation. Possible
future steps would be to establish queries of
ESSENCE data for Salmonella, as well as other
enteric disease diagnoses, and develop a protocol
for investigation of identified cases as a method
for earlier identification in the context of an
outbreak with ongoing exposures.
Figure 1 Number of Laboratory Confirmed Cases of
Salmonella Saintpaul by date of illness onset
United States, 2008 (Reference CDC)
For the 2008 outbreak of Salmonella Saintpaul
infections, a query was developed to search for
patients with chief complaints related to
Salmonella or tomatoes during April - August 2008
and to determine whether these individuals were
laboratory confirmed cases. The query contained
the search terms Salmon and Tomato. This
query enabled identification of chief complaints
for Salmonella that were misspelled, while also
being specific enough to reduce unwanted noise in
the query results. In addition, a retrospective
query was developed to search for chief
complaints of Salmonella from January 2007 -
April 2008 and to determine if these individuals
were laboratory confirmed cases. This query also
used the search term Salmon. Follow up
investigations of these potential cases were
conducted to obtain information on patient
discharge diagnosis and laboratory results, and
if applicable, length of time for case reporting
and identification at various points in the
reporting process. Cases identified using
ESSENCE were compared to notifiable diseases
reports.