The Role of Epidemiology in Outbreak Prevention and Control

1
The Role of Epidemiology in Outbreak Prevention
and Control

• Martin Cilnis MPH, MS
• Outbreak Prevention and Control Section
• Tuberculosis Control Branch
• California Dept of Public Health
• February 27, 2008

2
Objectives

• Describe the California TB Control Branchs
  Outbreak Response Team (ORT)
• Outbreak epidemiologists role
• Resources available for local jurisdictions
• Examples of epidemiologic assistance
• Using genotyping in outbreak detection and
  monitoring

3
CDPH TB Outbreak Response Team (ORT)

• ORT Public health nurse, physician, communicable
  disease (CD) investigator, CD manager,
  epidemiologist
• A resource for local TB programs
• Wide range of services available
• In 2007
• Provided assistance to 16 counties for
  investigation of 23 events
• 9 suspected or confirmed outbreaks, 11 extended
  contact investigations

4
The Epidemiologists Role

• Technical assistance to state/local TB programs
• Receive initial reports of outbreaks
• Analysis and interpretation of data
• Provide data collection and management tools
• Disseminate, communicate, and present data
• Outbreak surveillance

5
ExampleEpi Assistance to a Local TB Program

• Contact Investigation at a School

6
Contact Investigation at a School

• Background
• Student with pulmonary TB
• Smear-positive, cavitary disease
• 8 month infectious period (during school year)
• Contacts
• School (students/faculty)
• Household, other
• Much local media attention

7
Assistance sought by LHD

• Contact investigation (CI) planning consultation
  and prioritization
• Data collection plan and tools
• CI database tailored for school setting
• Real-time data analysis of contact evaluation
  results
• Consultation about if and how to expand CI

8
Activities by LHD (Partial List)

• Parent, student, and staff education
• Find and evaluate social and household contacts
• Place and read TSTs on campus
• Main cohort had at least 1 class with index
  case
• Expansion cohort persons in period after index
  cases class
• Data collection
• Enter data into an Excel database

9
Testing the Main Cohort

• Contacts with at least 1 class with the case

10
Main Cohort TST Results

• Higher than expected of TST positives

11
CI was expanded

• Contacts with less exposure were evaluated

12
Expansion Cohort TST Results

• Less evidence of transmission

13
Outcomes

• CI was expanded due to higher than expected TST
  () rates in the main cohort
• TST () rate of expansion cohort was close to
  expected background rate CI was not expanded
  further
• Conclusions based on data were used by county
  Public Information Officer as media talking points

14
Outcomes

• Genotyping
• The case shared the same genotype with about 35
  other cases in CA
• No known links to the other cases
• The vast majority of cases in the cluster are
  from the Philippines which suggests that this is
  a common strain there (the student case is
  African-American)
• Question Any links to the Asian community?

15
Using Genotyping Data In Outbreak Control
16
Genotyping In Outbreak Control and Prevention

• Genotyping data is extensively used by the ORT
• Confirm or refute outbreaks
• Characterize clusters
• Identify previously undetected outbreaks
• Evaluate efficacy of outbreak containment efforts
• Analysis of genotyping data with demographic and
  clinical patient data

17
A Brief Overview of the Genotyping Program
18
National Genotyping Project

• Program started in 2004
• 2 genotyping labs California and Michigan
• LHJs have the option of sending
• All culture () isolates (Universal)
• Some of the culture () isolates (Selective)
• No isolates

19
Genotyping Methods

• PCR-based methods
• Spoligotyping (Spacer Oligonucleotide Typing)
• MIRU (Mycobacterial Interspersed Repetitive
  Units-Variable Number of Tandem Repeats)
• RFLP-based method
• RFLP IS6110 (Restriction Fragment Length
  Polymorphism IS6110)

20
Spoligotype and MIRU

• Pro
• Rapid turnaround (2 weeks)
• Results easily read and compared
• Con
• Not as powerful as other commonly used methods

21
Spoligotype and Miru Results
22
RFLP Method
Done for isolates that match by Spoligotype and
Miru

• Pro
• High discriminatory power if 6 bands
• Slow turnaround (3 weeks)
• Con
• Results are difficult to compare across sites

23
ExampleUsing Genotyping in an Outbreak
Investigation

• Cluster of TB Cases with HIV/AIDS

24
HIV/AIDS Cases

• Increase number of TB cases with AIDS in a
  jurisdiction
• 4 in 2006, 13 in 2007
• Cases were U.S.-born and most have a history of
  substance abuse

?Is this an outbreak?
25
Investigation Plan

• Review case charts to uncover epi links
• Interview patients to find additional contacts
  and transmission settings
• Examine genotyping and surveillance (RVCT) data
• Obtain lab specimens for all cases for genotyping
• Send a notification to surrounding LHDs about
  outbreak

26
Genotypes of the Cases

• 6 cases had the same genotype
• Two of the cases had known links
• 8 cases in different genotype clusters
• 2 cases with unique genotypes

27
Outbreak Case Characteristics

• Characteristics of the cases
• INH-resistance
• Drug use
• HIV()/AIDS
• African-American race/ethnicity
• Developed case definition for outbreak
• 8 additional cases had the same genotype
• 6 were reported in other counties, however, no
  links were found

28
Links to Previous Cases

• County staff suspected links to previous cases
  reported in 1990s
• Searched genotype database from national
  genotyping project 1996-2000
• Genotype matches found
• 10 cases reported before 2004
• Some similarities to recent cases
  African-American race/ethnicity, INH-R, homeless
  or substance abuse

29
Summary

• Genotyping was an important tool
• Helped focus the investigation
• Confirmed that this strain has been present in
  this community for over 10 years
• Four previously unknown links among the cases
  were found
• Suspected links involving drug-use
• Re-interview of 6 cases identified
• gt25 additional contacts requiring evaluated
• Potential settings of transmission (HIV Center
  attended by 4 cases - contacts were tested and
  evaluated)

30
Genotype Cluster Surveillance

• Detecting outbreaks

31
Outbreak Detection

• Goal Early detection of outbreaks
• Genotype-surveillance (RVCT) data
• Outbreak flags
• Time a spike in clustered case over a period of
  time
• Location e.g., genotype cluster in one
  jurisdiction
• Case risk factors/high risk settings e.g.,
  homelessness, substance abuse, HIV()
• Limitations Common strains, resources of
  jurisdictions

32
Outbreak Detection Methods

• Screen clusters using outbreak flags
• Discuss results with ORT and categorize the
  selected clusters 1 (very concerning notify
  LHD) to 4 (monitor cluster)
• Notify LHD and send cluster reports
• Follow-up with LHDs
• Is the genotype cluster an outbreak?
• If an outbreak, was it already known by the LHD?
• Steps taken to halt transmission?
• Assistance required from state ORT?

33
Genotype Cluster Report
34
Outbreak Cluster Assessment

• 58 clusters were flagged (out of gt500 clusters in
  CA)
• ORT identified 26 clusters as concerning or
  possibly significant
• 7 very concerning and LHD was contacted to alert
  of possible outbreak
• 3 were somewhat concerning and an FYI was sent to
  LHD
• Monitor 14 clusters

35
Outbreak Detection Results

• Preliminary outcomes
• The very concerning clusters were known by LHDs
• So far, no previously unknown links found
• One cluster is currently being investigated

36
Resources and Information

• CDPH TB Outbreak Response Team
• http//ww2.cdph.ca.gov/programs/tb/Pages/TBSurvFor
  msTBCB.aspx (under Outbreak Response and
  Reporting section)
• Genotyping
• CDC Tuberculosis Genotyping Guide
• http//www.cdc.gov/tb/genotyping/manual.htm
• Genotyping Program in California
• http//ww2.cdph.ca.gov/programs/tb/Pages/TBSurvFor
  msTBCB.aspx (under Genotyping Isolate
  Submission and Instructions section)

37
Questions?

• Martin Cilnis
• [phone]
• mcilnis_at_cdph.ca.gov