STS-CCAS Database Update
By David Vener, MD
Texas Children’s Hospital
STS-CCAS Database Update (Sept. 2011)
Our database efforts went “live” on January 1, 2010 and if our experience at Texas Children’s is similar to others, then “rocky” is a good adjective. However, everyone involved in this effort knew that our first year(s) were going to have a steep learning curve and in that sense it has gone exactly according to plan!
The STS has made tremendous strides towards achieving almost 100% participation for their surgical partners: 98 programs (96 from within the U.S. and one each from Canada and Japan) have returned signed contracts for the 2010 data harvest, and 16 have included their anesthesia colleagues in the process. This is out of 122 programs in the United States that provide Congenital Heart Surgical Services.
Our anesthesia data will be harvested in Springtime 2011. Once we have a look at it, the database committee will evaluate how and if we want to analyze these initial data points or if we want to produce more of a narrative document pending sufficient patients being reported. I suspect that many of the 16 programs planning to submit anesthesia data will have incomplete records for 2010 due to how quickly various software vendors rolled out their products and how effectively and comprehensively data entry has occurred at individual sites.
As was made clear from the very beginning of this process, this is a multi-year effort that will take some time to show results worthy of publication and dissemination. That being said, we need to start thinking about our next version. The STS works on a three-year version cycle and we will begin working on the next round of the database in July 2011. I strongly encourage anyone who has begun to work with the data fields to provide me and the committee with feedback about what works, what doesn’t work and what we want to add/delete/change. The earlier we get this information, the better it will be and the more likely to be incorporated into the next version.
On another note, one of the key components of the database for both the surgeons and, to a lesser extent, anesthesia is the development of the Unique Patient Identifier, which will allow the same patient to be tracked in a confidential manner across both space (institutions) and time. By submitting Protected Health Information (PHI) such as the patient’s name, parental information, gender, social security number, birth date and other identifying information to the data warehouse, the PHI is processed through a secure algorithm that produces as its end result a HIPAA-compliant non-traceable Unique Patient Identifier that can then be used to “marry” information coming from disparate sources to the same patient.
The more PHI data that is submitted, the stronger the likelihood of successfully matching the patient to the information. For example, a UPI generated only from the patient’s last name and birthdate would be much more likely to have duplication than one that included the parents' information and the patient’s gender and social security. Since many of our patients are either too young to have social security or are not US citizens, the use of that information is fairly limited as an identifier but can be very helpful for long-term mortality calculations.
Many IRBs are under the mistaken impression that HIPAA regulations prevent the transmission of patient identifying information to a secure data warehouse such as the Duke Clinical Research Institute. In fact, the very opposite is true. The HIPAA legislation includes language recognizing the importance of these efforts and specifically allows for the creation of data records oriented towards patient safety and outcomes.
Jacobs et al published an editorial in 2008 outlining the STS’s approach to the UPI and why it is so important, including out-of-hospital morbidity tracking through the National Death Index, the Social Security Death Master File as well as multi-institutional follow-up.1 This sort of longitudinal information has the potential to significantly impact the long-term results available to cardiologists, surgeons and families in their decision making processes. The STS web site has documentation available that discusses issues surrounding the UPI and HIPAA.2
If your organization or department is having trouble with IRB approval for using PHI, please follow up as soon as feasible. The legal documents are readily available and it will significantly impact the value of all of our efforts over time.
Please feel free to contact me directly at any time with questions or if your organization would like a copy of the paper record that we are currently using. Anyone wishing to participate in the database committee is encouraged to attend the CCAS Annual Meeting, held in conjunction with the SPA-AAP Winter Meeting, at the Sheraton San Diego from March 31 – April 3, 2011.
- Jacobs J, Haan C, Edwards F, et al. The Rationale for Incorporation of HIPAA Compliant Unique Patient, Surgeon, and Hospital Identifier Fields in The STS Database. Ann Thorac Surg2008; 86: 695-8.
- http://www.sts.org/documents/pdf/DCRI_FAQ-Technical_Document_re_Identifiers_in_Database_Final.pdf
