County of San Diego

1
Using Syndromic Surveillance Data for Enhanced
Case-Capture of Conditions of Public Health
Importance
International Society for Disease
Surveillance October 11-12, 2007
Jeffrey Johnson Brit Colanter Azarnoush
Maroufi Jennifer Nelson Michele Ginsberg
County of San Diego Health Human Services Agency
2
OBJECTIVE
Assessment of a syndromic surveillance system
that identifies - Legally reportable diseases
- Conditions of public health interest
County of San Diego Health Human Services Agency
3
SAN DIEGO COUNTY

• Population 3 million
• International border
• Large military bases
• Biotechnology Hub
• 21 Emergency Departments

4
LEGAL REPORTING PROCESS

• California Code of Regulations
• Title 17-Public Health
• Health care providers are required to report
• Section 2500
• Over 80 different diseases to the local health
  department
• Section 2503
• Suspected outbreaks and
• unusual occurrence
• of disease

5
Emergency Department Syndromic Surveillance in
San Diego County

• 8 Hospitals (42 of all County ED visits)
• Daily electronic files
• Data Fields
• Date/time, chief complaint, age, zipcode,
  disposition, ED diagnosis, mode of arrival
• Syndrome groups 11 core groups
• Rash, respiratory, GI, chest pain, etc.
• Automated data processing, analysis and summary
  of results
• Review and response according to protocols

6
ED Syndromic Surveillance Data Processing
Hospital
Follow-up Response
Individual Hospital Analysis
Public Health FTP Site
Syndrome Grouping Process
Aggregated Hospital Analysis
Daily Data File
Individual case (PICC List) Process
Listing of detected cases
Data Processing
7
Priority Infection Case Capture (PICC)

• Text-mining for key words
• Diseases/conditions of public health importance
• Query run daily with automated email results

Chief Complaint Diagnosis fields merged
Reportable conditions
Non-reportable conditions of interest
8
PICC Key Diseases/Conditions

• HUS
• INFLUENZA (FLU)
• KAWASAKI
• LEAD
• LEGIONELL
• LEPTO
• LISTERIA
• MALARIA
• MEASLES
• MENING
• MRSA
• MUMPS
• OUTBREAK
• PERTUSSIS
• PESTICIDE
• PLAGUE
• Q FEVER

• RABIES
• RADIATION
• SALM
• SARS
• SCOMBROID
• SHIGELLA
• SMALLPOX
• TB
• TETANUS
• TOXO
• TRICH
• TSS
• TULAREMIA
• TYPHOID
• VIBRIO
• WEST NILE
• WNV
• YERSINI

• ANTHRAX
• BOMB
• BOTULISM
• BRUCELL
• CAMPY
• CHEMICAL
• CHOLERA
• COCCI
• CRYPTO
• DENGUE
• DRUG RESISTANT
• E COLI
• ENCEPHALITIS
• FOOD POIS
• FOODBORNE
• GIARD
• HANTA
• HEPATITIS

9
STUDY DESIGN

• Prompted by a variety of real detected cases or
  events
• Retrospective Assessment of PICC results
• Descriptive analysis of PICC cases
• 2006 data from participating hospitals
• Prospective Assessment
• Meningitis related admissions
• Follow-up of potential cases detected by PICC
• 19 weeks 10/23/2006 4/5/2007

10
RESULTS Retrospective Assessment
Total ED Encounters (2006) 312,334 100
Occurrences of potential 222
.07
reportable diseases
Measles Meningitis Mumps Pertussis
Salmonella Scombroid Shigella Tetanus
Toxic Shock Syndrome
Toxoplasmosis West Nile Virus
Botulism 1 0.5 Campylobacter 1
0.5 Coccidioidomycosis 3 1.4 Cryptosporidium
1 0.5 Cysticerosis 2 0.9 E.
Coli 1 0.5 Encephalitis
11 5.0 Giardia 1 0.5 Hepatitis A 5
2.3 Lyme 3 1.4 Malaria 3 1.4
2 0.9 163 73.4 2 0.9 6 2.7 1
0.5 3 1.4 3 1.4 5 2.3 2
0.9 1 0.5 2 0.9
not yet confirmed as true cases
11
STUDY DESIGN - Prospective Review of Meningitis

• Criteria for follow-up
• Meningitis within chief complaint
• Patient was admitted
• Study period 19 weeks (Oct 2006-April 2007)

12
RESULTS - Prospective Review of Meningitis
Traditional reporting of confirmed/reported
meningitis cases reported during study period.
N137
of confirmed/reported meningitis cases from
EDs participating in Syn Surv. N75
PICC of meningitis cases detected by PICC
during study period. N36
62 Cases
23 Cases
13 Cases
52 Cases
False (4) or unknown (9)
Confirmed reported
13
RESULTS - Prospective Review of Meningitis
Confirmed Reported 12 (75) 4 (67) 4 (50) 2
(50) 1 (100) na na 0 (0) 23
False or Unknown 4 (25) 2 (33) 4 (50) 2
(50) 0 0 0 1 (100) 13
Hospital A B C DEF G H Total
PICC 16 6 8 4 1 0 0 1 36
14
RESULTS - Prospective Review of Meningitis
Assessment of Timeliness Time until Public
Health awareness
Mean 3 days Range 1-6 days
PICC Detection
On average, PICC resulted in a 2 day earlier
notification
Traditional Disease Reporting
Mean 5 days Range 1-15 days
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
15 16 17 18
DAYS
15
RESULTS - Prospective Review of Meningitis
Summary of PICC detected meningitis cases that
were confirmed and reported (n23)

• PICC detected first before traditional
  reporting 6 (26)
• Final Diagnosis Bacterial Mening. 1 (4)
• Viral Mening. 13 (57)
• Fungal Mening. 3 (13)
• Unknown/Other 6 (26)
• Referred for WNV Testing 19 (83)
• Specimen sent to PH Lab 17
• Etiologic agent Enterovirus 3 (13)
• Cryptococcus 3 (13)
• Prob Strep Pneu. 1 (4)
• Unknown 16 (70)

16
RESULTS - Prospective Review of Meningitis
Follow-up Process

• Public Health Staff resources
• Surveillance Epidemiologist
• Review PICC results, check if case reported,
• Communicate case(s) to Epi investigator Epidemi
  ologist Investigator
• Contact hospital. Review lab results and
  clinical history
• Time spent
• Public Health Approximately 10 minutes per case
• Hospital ??

17
SUMMARY

• Benefits
• Identification of individual cases
• Earlier reporting of cases
• Text-mining techniques can be easily modified to
  query for items of interest

• Limitations
• Pre-diagnostic false cases
• De-identified data limits ability to follow-up
  on all cases
• More assessment needed to understand true
  sensitivity and specificity
• Hospital staff time and resources for follow-up

18
CONCLUSIONS

• Syndromic surveillance can be a useful tool to
  detect individual cases of public health
  importance
• Flexibility of the system adjustment to fit
  local needs
• Need data identifiers to make follow-up on cases
  more feasible on an ongoing basis
• Further evaluation is needed to assess public
  health control measures initiated by earlier
  detection of cases
• Redundant surveillance systems ensure more
  complete reporting

19
THANK YOU For More Information Jeff
Johnson 619-531-4945 Jeffrey.johnson_at_sdcounty.ca.g
ov