Infection Control Surveillance Technologies

1
Infection Control Surveillance Technologies

• Emily Sickbert-Bennett, MS, CIC
• UNC Healthcare System

2
Surveillance

• CDC
• The ongoing systematic collection, analysis,
  and interpretation of health data essential to
  the planning, implementation, and evaluation of
  public health practice, as well as the timely
  dissemination of these data to those who need to
  know.

3
History

• 1958 nationwide S.aureus infections in hospitals,
  AHA recommended hospitals implement nosocomial
  infection surveillance
• 1960s CDC recommended surveillance part of
  Hospital Infection Control programs
• 1976 TJC first included IC surveillance standards
  in requirements for accreditation
• SENIC 1985 showed hospitals with strong
  surveillance programs with prevention and control
  measures improved nosocomial infection rates

4
CDC MMWR 50, RR-13, 2001
5
Surveillance Types

• Full house surveillance
• Targeted surveillance
• Syndromic surveillance
• Automated surveillance

6
Full House Surveillance

• Attempts to identify all healthcare-associated
  infections (HAIs) in all patients at all times
• Establishes baseline data for all locations and
  services
• Provides monthly attack rates for feedback to
  clinicians

7
Targeted Surveillance

• Looks only at a segment of the patient population
  with the highest risk of disease (e.g., ICU)
• Surveillance components are selected by the IP
  and the surveillance time period can be variable
• In addition to surveillance for HAIs, may include
  surveillance for specific pathogens (e.g., VRE)

8
Location of Respiratory Infections (Medicine and
Surgery units 2004-5)
44 of respiratory infections occurred outside ICU
ICHE, 2007, 281361
9
Location of Urinary Tract Infections (Medicine
and Surgery units 2004-5)
77 of UTIs occurred outside ICU
ICHE, 2007, 281361
10
Location of Bloodstream Infections (Medicine and
Surgery units 2004-5)
ICU
Non-ICU
68 of all BSIs occurred outside ICU
ICHE, 2007, 281361
11
What is Syndromic Surveillance?

• Surveillance using health-related data that
  precede diagnosis and signal a sufficient
  probability of a case or an outbreak to warrant
  further public health response
• -CDC

12
NC DETECTwww.ncdetect.org

• FREE Surveillance tool created by the North
  Carolina Division of Public Health/UNC Department
  of Emergency Medicine in 2004
• Data from
• Hospital Emergency Departments
• Carolinas Poison Center
• Pre-hospital Medical Information System
• Piedmont Wildlife Center
• NCSU College of Veterinary Medicine Laboratories
• Urgent Care Centers (coming soon)

13
Automated Surveillance

• Systematic application of medical informatics and
  computer science technologies for infection
  control surveillance
• Surveillance Steps
• Collection
• Analysis
• Interpretation
• Dissemination

14
Data collection
Traditional Sources Data Entry
Automated
EMR
EMR
15
Automated Data Collection/Integration

• Numerator (Infections)
• Lab information systems
• Admission Discharge Transfer databases
• Online Medical Records
• Denominator
• Census data admissions and/or patient days
  (denominator)
• Device days (denominator)
• Surgery procedure specific data (denominator and
  risk adjustment)

16
Data analysis/interpretation
Traditional
Automated
17
Automated Data Analysis/Interpretation

• Data mining using technology to discover
  patterns and relationships that then classify
  data
• Query based data management requires user input,
  does not seek patterns independently

18
Data dissemination
Traditional
Automated
19
Automated Data Dissemination

• Create dashboards
• Schedule reports to be sent on a regular schedule

20
Advantages

• Traditional
• Inexpensive technology
• Can manually aggregate data not available
  electronically
• Benchmark to own historical rates
• Automated
• Requires fewer hours of IP time on surveillance
• Integrates data available electronically
• May simplify reporting requirements State or
  NHSN
• Generate more reports, more quickly automated
• Generate alerts for events of significance

21
Disadvantages

• Traditional
• Very time consuming
• For case finding, case investigation
• For data analysis
• For report dissemination
• Automated
• Expensive to implement (especially initially)
• May lose some autonomy over application of case
  definitions

22
Evaluation Characteristics

• Simplicity
• Flexibility
• Data Quality
• Acceptability
• Representativeness
• Timeliness
• Stability

--CDC MMWR 50, RR-13, 2001
23
Simplicity

• Simplicity of structure and ease of operation
• Measures
• Number of data sources reporting, receiving data
• Methods of data collection, data management, data
  analysis and dissemination
• Amount and type of data required for cases

24
Flexibility

• Adaptable to changing information needs with
  little additional time, personnel, allocated
  funds
• Measures
• Retrospective observation of how system has
  responded to revised case definition, additional
  data sources, new information technology

25
Data Quality

• Completeness and validity of the data
• Measures
• Percentage of missing data
• Compare to true traditional surveillance methods

26
Acceptability

• Willingness of persons and organizations to
  participate in surveillance system
• Measures
• Completeness of data
• Timeliness of data reporting

27
Representativeness

• Accurately describes the occurrence and
  distribution of events in the population by place
  and person
• Measures
• Comparison of characteristics from reported
  events to actual events with traditional
  surveillance methods

28
Timeliness

• Speed between steps in a surveillance system
• Measures

29
Stability

• Reliability and availability of the system
• Measures
• Number of unscheduled outages
• Costs involved with repair
• Percentage of time system is fully operating
• Desired and actual amount of time for system to
  collect, receive, manage, and release data

30
Traditional vs. Automated Surveillance
31
Strategies

• Create wish list
• Describe data flow to identify opportunities for
  automation
• Engage stakeholders
• Utilize APIC tools
• Prepare Business Case

32
UNC Example

• Wish List
• Fully automated collection of all data available
  electronically
• Ability to still perform case review and manually
  apply CDC case definitions
• Customizable queries and reports
• Automated tools for cluster/outbreak investigation

33
Data Flow
Possible Cluster/Outbreak investigations (e.g.,
timelines, susceptibility patterns)
Alert to significant infections
34
Engage Stakeholders

• Who?
• Information Technologists
• Lab and System wide
• Infection Control
• Lab
• Pharmacy
• Performance Improvement
• Administration
• How?
• Share tools, develop questions for vendors
• Attend live or web demos

35
APIC tools

• APIC Starter Questions to identify vendor(s) to
  use more extensive assessment tool
• APIC Surveillance Technology Assessment Tool
• Informatics Glossary Reference
• Each other!

36
APIC tools
37
Business Case

• Vendor may provide cost/benefit analysis
• Be prepared to answer administrators questions
  of how many IP positions can be eliminated?
• Involve other departments (lab, pharmacy) that
  may gain benefit from system and provide cost
  sharing

38
Thank you.