Transcript Lessons Learned from the Society of Thoracic Surgeons (STS)

Lessons Learned from the Society of
Thoracic Surgeons (STS) Congenital
Database
September 25, 2015
Robert J. Dabal, MD
Associate Professor of Surgery
I
do not have any relevant
financial relationships with any
commercial interests to disclose.
 While
not an expert on the
subject, I did serve on the STS
Congenital Surgery Database
Task Force for about 4 years.
Overview

History of the STS database

Applications
Quality Improvement
Research

Limitations

Future Directions
History of the STS Database

Founded in 1989

Quality improvement and
safety initiative of the STS

Adult cardiac and thoracic,
followed by congenital

Joint effort of the STS
Congenital Database Task
Force and the Duke Clinical
Research Institute (DCRI)
History of the STS Congenital Database

Congenital database started in 2002

116 North American congenital participants
representing 127 hospitals
124/125 US hospitals, 3/8 Canadian
hospitals
History of the STS Congenital Database
History of the STS Congenital Database
History of the STS Congenital Database
History of the STS Congenital Database

Database structure
One harvest per year from 2004 to 2007
Two harvests per year beginning in 2008 –
Spring and Fall

23rd harvest – today

4 year span – July 1, 2011 to June 30,
2015
• 149,780 operations in last harvest
Quality Improvement

How do we compare?
Mortality
Patient mix/
preoperative factors
Operative times
Length of stay
Complications
Quality Improvement

Data summary

Risk adjusted mortality

Participant specific
outcomes (benchmarks)

Lesion specific outcomes

Missing data summary
Quality Improvement

Data Summary
 Patient and procedure volume
 Discharge mortality
 Procedural complexity

By participant center

By age groupings
 Frequency of all diagnoses and
procedure types
Quality Improvement

Risk-adjusted
mortality
 UAB/Children’s of
Alabama 2015

250 index operations
(high volume center)

5 total deaths

2% overall mortality,
STS 3.3%

STAT, RACHS-1
Quality Improvement

Patient specific
outcomes
(Benchmark
operations)
Quality Improvement

Lesion specific
outcomes
 Demographics
 Preoperative factors
 Non-cardiac
abnormalities
 Operative information
 Complications
 Mortality
Quality Improvement

Missing data
Research

Initial publications in April 2000

Total of 65 publications over 15 years
 Descriptive (43)

Methodology, risk stratification
 Comparative, linkages (9)
 Lesion, operation or patient specific outcomes (12)
Research

Descriptive (43)
 Mostly early papers - methodology
 Later papers on definitions, clarifications and risk
stratification
Research

Comparative, linkages (9)
Other databases, social security, CHSS
Research

Congenital Heart
Surgeon’s Society

Founded in 1985 by
Dr. John W. Kirklin and
Dr. Eugene Blackstone

Data center housed at
UAB for first decade

Dr. James K. Kirklin,
current research chair
Research

Lesion, operation or patient specific outcomes
(12)
Limitations

Definitions change – moving target
Spring Harvest – changes in operative
mortality from hospital discharge to database
discharge
Fall Harvest – new methodology for
determining primary procedure – highest
STAT score

Retrospective to include all 4 years
Limitations

Missing data
Greater than 10% missing data excluded
Approximately 10% audited per year
Limitations

Only 50 of 116 programs
participate in the Congenital
Cardiac Anesthesia Society
(CCAS) module
Over 64,000 anesthetic
records

Anesthesia tracking is not
available for procedures done
by providers with IDs
Limitations

Voluntary public
reporting began
in January 2015
Available only
for programs,
not surgeons

Opt-in not
available until
Spring 2016
Limitations

Public reporting

8 variables in the mortality risk model – age group,
primary procedure, weight, prior cardiothoracic
operation, non cardiac abnormalities, chromosomal
abnormalities, prematurity and 6 preoperative factors
Future Directions

More robust data collection

Higher rate of audits

Linkage to other databases – PC4, ACC

Mandatory participation

Better system of public reporting
Limited rankings
Proof of safety – standard of care
Summary

The STS congenital database is a valuable
resource for clinicians in both the areas of
quality improvement and research.

The cumulative experience is growing
dramatically and will continue to do so over the
next decade.

A successful program can not be evaluated
solely on the basis of the STS congenital
database.
Thank you for your attention