Ventilator-Associated Event Surveillance:
CDC Updates
Shelley S. Magill, MD, PhD
July 16, 2015
National Center for Emerging and Zoonotic Infectious Diseases
Division of Healthcare Quality Promotion
Overview
Review the current status of National Healthcare Safety
Network (NHSN) Ventilator-Associated Event (VAE)
surveillance
Discuss VAE data reported to NHSN in 2013-2014
Provide definition and IT updates
STATUS OF NHSN VAE SURVEILLANCE
VAE
Ventilator-Associated
Condition (VAC):
Based on changes in PEEP and
FiO2
Infection-related
Ventilator-Associated
Complication (IVAC):
VAC with general evidence of
infection
Possible VentilatorAssociated Pneumonia
(PVAP):
IVAC with lab evidence
of pneumonia
VAE Surveillance Timeline
VAE
PEEP criterion Changes to
implemented changes
eligible
in NHSN
antimicrobials
1/2013
8/2013
1/2014
Additional
organism
exclusions
CMS reporting
requirement for
LTCH QRP
1/2015
1/2016
Consolidation of Addition of new,
Additional
Possible and
optional
guidance for
microbiological Probable VAP into denominator
single “PVAP”
(“EMV”)
criteria
definition
VAE DATA REPORTED TO NHSN
Facilities Reporting In-Plan VAE to NHSN, 2013-2014*
4%
1615 facilities in 2013
1893 facilities in 2014
2017 facilities in 2013
and/or 2014
N=2017
6%
7%
General
hospital
LTAC hospital
83%
Critical access
hospital
Other
*Includes any facility indicating in-plan VAE surveillance in 2013 or 2014, regardless of whether data submission is complete for all locationmonths of reporting. Preliminary and subject to change.
Locations Reporting In-Plan VAE Data to NHSN,
2013-2014*
N=3927
Total of 68,695
location-months
of VAE
surveillance data
reported in 20132014
3927 unique
locations
Med/surg ICU
2% 2%
Medical ICU
10%
Cardiothoracic ICU
3%
Med/surg ward
4%
42%
Cardiac ICU
4%
Surgical ICU
5%
Stepdown
LTAC ward
5%
Neurosurg ICU
6%
6%
11%
Medical ward
Trauma ICU
Other
*Includes any facility indicating in-plan VAE surveillance in 2013 or 2014, regardless of whether data submission is complete for all locationmonths of reporting. Preliminary and subject to change.
Selected Critical Care Location VAE Rates, 2014
**preliminary**
VAEs per 1000 ventilator days
Location
LTAC
No. units*
18
Pooled
mean
2.00
Med/surg
1478 (1325)
5.60
0
3.09
11.51
Cardiothoracic
232 (231)
6.16
0
4.93
12.79
Cardiac
185 (181)
6.41
0
5.45
13.61
36 (34)
6.55
0
6.52
18.07
Medical
383 (354)
7.37
0
5.17
13.06
Surgical
186 (183)
7.71
0
6.28
15.73
99
8.29
0
6.66
19.44
83 (82)
11.79
2.07
10.95
21.45
Burn
Neurosurgical
Trauma
10%ile
--
50%ile
--
90%ile
--
Data are preliminary, unpublished, subject to change. *Unit no. in parentheses are those reporting >50 vent days per year and included in
rate distributions.
Distribution of VAE Specific Sites (N=35,040)
Po/Pr VAP,
16%
“IVAC-plus”
IVAC,
21%
Preliminary, unpublished data, subject to change.
VAC only, 63%
Criteria Used to Report VAEs (N=35,038)
Both PEEP
and FiO2,
10%
FiO2 only,
22%
PEEP only,
68%
Preliminary, unpublished data, subject to change.
Excludes 2 events with incomplete criteria reporting.
What’s the breakdown of VAEs in different ICUs?
80%
Proportion of VAEs
70%
VAC
67%
65%
65%
IVAC-plus
60%
60%
59%
59%
50%
40%
40%
33%
35%
Cardiac
Medical
41%
41%
56%
44%
54%
46%
35%
30%
20%
10%
0%
Med/surg
Preliminary, unpublished data, subject to change.
Surgical Cardiothor Neurosurg
Burn
Trauma
Time from Intubation/Mechanical Ventilation
Initiation to VAE (N=35,021)*
10000
35%
39%
9000
25%
Number of VAEs
8000
7000
19%
16%
6000
13%
5000
4000
9%
7%
3000
6%
2000
4%
1000
0
Day 3
Day 4
Day 5 Day 6 Day 7 Day 8
Mechanical Ventilation Day
Preliminary, unpublished data, subject to change; *19 records excluded due to data entry error.
Day 9 Day 10+
Time from Hospital Admission to VAE, Among Patients
with VAEs on Mechanical Ventilation Days 3-4
(N=12,361)*
60%
4500
32%
4000
Number of VAEs
3500
28%
40%
3000
2500
2000
15%
10%
1500
1000
6%
4%
500
3%
2%
Day 8
Day 9
0
Day 3
Day 4
Day 5
Day 6
Day 7
Hospital Day
Day 10+
Preliminary, unpublished data, subject to change. *One record excluded due to data entry error; 2 records excluded due to missing data.
Timing of VAE Onset
VAEs in patients in
the hospital <5 days
AND on the
ventilator <5 days
= 32% of all VAEs
VAEs in patients
in the hospital OR
on the ventilator
for at least 5 days
= 68% of all VAEs
Preliminary, subject to change.
Summary
Majority of VAEs occur on/after mechanical ventilation day 5
or hospital day 5
Suggests most are healthcare-associated
Distribution of VAC vs. IVAC-plus differs by ICU type
Reflects capture of a variety of conditions that may be more or less
common depending on the patient population
Additional work needed to:
Describe VAEs in different patient populations
Understand reasons for differences in VAE specific event distributions
Understand clinical correlates of early-onset VAEs
Consider whether VAC definition can be modified to exclude events
where oxygenation has not worsened but rather PEEP is being
increased to allow for reduction in FiO2
VAE DEFINITION AND IT UPDATES
Definition Updates
Minor changes for 2016—adding recently approved
antimicrobial drugs to the list of drugs eligible for meeting
IVAC and PVAP definitions
Pediatric VAE
Recent work exploring potential pediatric VAE-like definitions
presented at SHEA 2015
Reconvening the work group in September 2015
Goal to implement in NHSN in 2017
Synthetic Data Set
An EHR vendor seeking to automatically detect VAEs within
their data will first read and interpret the definitions from
written description/VAE protocol
Need a way to electronically confirm that the vendor is
applying the VAE definitions accurately
Creation of a synthetic data set
Modified from slide courtesy of Cindy Gross and Barry Rhodes
Use of Synthetic Data Set
Two XML files available to vendors:
XML File 1 : Sample patient records that have embedded in
them, various VAE event types
Vendors import sample set into their system and run their
implementation of the VAE algorithm against it
The data elements and definitions map directly to the VAE protocol
which may be found here.
http://www.cdc.gov/nhsn/PDFs/pscManual/10-VAE_FINAL.pdf
XML File 2: The result set with the expected VAE
determinations
Vendors compare the results their algorithms generate against the
result set
Slide courtesy of Cindy Gross and Barry Rhodes
Web Service
A service for vendor systems to submit de-identified data
automatically or for individual users to submit de-identified
datasets through a web page
XML or CSV file format
Data elements
Patient index (de-identified)
Ventilator day
Daily minimum PEEP
Daily minimum FiO2
Min temp / max temp
Antimicrobial agent
Assumes daily min PEEP and FiO2 are determined accurately
Slide courtesy of Cindy Gross and Barry Rhodes
Current Status
Synthetic data sets:
Provided to the vendor community
Available upon request to [email protected]
Web service
Available on request for beta testing
Working to move into a production environment in fall 2015
Slide courtesy of Cindy Gross and Barry Rhodes
Acknowledgments
DHQP staff, including Qunna Li, Cindy Gross, Barry Rhodes,
Jonathan Edwards, Maggie Dudeck, Kathy Bridson, Susan
Hocevar
NHSN facilities and users
VAP/VAE Surveillance Definition Working Group and
Neonatal/Pediatric Working Group
CDC Prevention Epicenters
Many subject matter experts
Federal partners
The findings and conclusions in this presentation are those of the author and do not necessarily
represent the views of the Centers for Disease Control and Prevention.
Questions?
[email protected]
For more information please contact Centers for Disease Control and Prevention
1600 Clifton Road NE, Atlanta, GA 30333
Telephone, 1-800-CDC-INFO (232-4636)/TTY: 1-888-232-6348
E-mail: [email protected]
Web: www.cdc.gov
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of
the Centers for Disease Control and Prevention.
National Center for Emerging and Zoonotic Infectious Diseases
Division of Healthcare Quality Promotion