| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1999;68:350-352
© 1999 The Society of Thoracic Surgeons
a St. Peter’s Hospital, Albany, New York, USA
Address reprint requests to Dr Dziuban, 319 South Manning Blvd, Suite 301, Albany, NY 12208
e-mail: sdziuban{at}albany.net
Presented at the Thirty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 26–28, 1998.
Abstract
Background. Clinical database information is in wide demand, but it is not always used to its full potential. Clinicians must learn to be the experts and to assert leadership in the use of their own data.
Methods. Clinical databases provide unique perspectives on the full process of care for a heterogeneous population of patients. They focus beyond individual providers, to their interaction within a complex system of care. Exploring questions that arise from this data can identify system issues, which are invisible to individual practitioners or specialties using other traditional review methods.
Results. Clinical database information helped our hospital staff identify problems in their approach to a small subset of high risk coronary bypass patients. Multiple system changes resulted in a dramatic reduction in mortality. Collateral impact on all care reduced overall mortality from 4.5% to below 2%.
Conclusions. The greatest opportunities for improvement in patient care often lie in the areas where specialties or teams interface, eg, in overlapping or transferring care. The whole system of care for each patient must be optimized, not just the individual specialty components. Clinical database information provides a way to evaluate and improve the overall process and results of the broader system of patient care.
Introduction of panel by STS Database Chairman, Frederick L. Grover, MD
As Dr Kaiser mentioned, we hope to bring the use of databases into their practical application, how you can translate the numbers that you get into improving your results, and how this can be helpful to our specialty. We have five discussants who will highlight various aspects of the practical application of databases.
We’re all swimming with sharks these days. The federal government, state governments, payers, managed care, consumer groups, lay press — everyone wants to review our care and our data. Our best option is to not only know our own data, but to also be effective in using it. We have to take leadership with our own data. After all, we are the experts in our own patient care, which is what our data describes.
I would like to focus specifically on how we can use our database information effectively, and the valuable and different information we can obtain from it. This will identify an important point about optimization of care. Then, I would like to take you vicariously through the experience of one cardiac surgery team, as they learned this process under trial-by-fire.
Database information has a life cycle. It begins with patient care, which is the basis for our data. We collect data and we build a model of our care in order to study it. Having this data is just the beginning. We then need reports to give us useful information. There are countless types of useful reports, but they are all only intermediate steps toward reaching our goal. Our real goal is the productive use of this information to modify and improve our care. Of course, this leads us back full-circle to the patient care, and so the process begins again. The fact that we use our database information in this way is actually more important than the numbers at any given point in time. A later example will illustrate this point.
What is special about clinical database information? It gives us a unique perspective not available from any other source.
First, it allows a scientific approach to mainstream patient care. Traditional studies tend to involve either a single case report, or a clinical trial in a homogeneous subset of patients. A clinical database allows us to study the entire heterogeneous mainstream of patient care, and then compare it to some external standard. This is a very powerful perspective and it can involve good science, although the tools are somewhat different from our usual ones.
Second, database information is unique because it spans the full process of care for all of these patients. For example in cardiac surgery it spans patient characteristics, the disease process, and all care through catheterization, anesthesia, surgery, and postoperative care, leading ultimately to the patient outcomes.
Third, database information is relevant to everyone involved, not just the surgeons. Advocates of the report card approach often oversimplify by reporting outcomes as if they were due to the surgeons alone. We must remember that many other specialties participate and influence our patient care and outcomes. We should therefore involve these others in reviewing and improving the care process.
This perspective emphasizes an essential fact. Integration among specialties is necessary to achieve optimization of care. Consider a situation where each specialty performs to optimize their own care: the best surgery, the best cardiology, the best anesthesia, and the best nursing. If each specialty is optimized separately, the overall process remains sub-optimized. Only if the various specialties join to form a single process can that process be optimized as a whole. In today’s world of highly specialized care, many of the opportunities for improvement and optimization of care occur at the interfaces and boundaries between specialties.
An example from real life experience [1] illustrates these principles. Imagine waking up one morning to read in the headlines, that your hospital has the highest cardiac surgery mortality in the state. This may crystallize some of the fears about what could occur if others were to use our data in inappropriately. This happened in late 1992, to one of the hospitals at which my group practices. The headlines were regarding coronary artery bypass graft (CABG) mortality. The hospital’s observed or actual mortality was 4.5%. The expected mortality, calculated from risk factors in the state database, was 2.2%, yielding an observed-to-expected mortality (O/E ratio) of about 2 to 1, meaning that the risk-adjusted mortality was about twice as high as the state average.
The surgeons involved were highly competent and well trained, just like you. Before this headline, a mortality of 4.5% seemed reasonable, given the number of high-risk cases involved. You can imagine reactions to this news — starting with the shock and disbelief that the data was correct. When the data was confirmed, suspicion fell next on the risk-adjustment method, questioning whether it accounted enough for the sicker patients. The whole situation was extremely frustrating; people were criticized for having a problem, but were not told exactly what it was or how to fix it.
Following a traditional approach, intensive case reviews were provided by many parties from different departments, and all of these came to the same conclusions. Most of the deaths had occurred in very high-risk patients, and no major errors or substandard care was found. No useful information had emerged to help clarify the problem, or offer a potential solution.
Things changed when we went back to the raw data, exploring the database for useful information and patterns, and compared our local figures in different ways to the New York State averages. Some surprising information emerged. In 95% of surgical cases, the mortality was lower than state average. The higher mortality was found concentrated in a small subset (5% of the cases). The overall care was excellent, but there was a problem in a small subset of high-acuity emergency cases.
Doing this kind of database exploration is actually quite simple. First, you need to have open access to the data in your database, so you can explore it with common software tools. We started out using a personal computer with statistical software, but later used desktop database software (Microsoft Access). You need to take a flexible approach, looking at the data from different angles. Comparable external data is necessary to put your findings in perspective. The Society of Thoracic Surgeons’ national database is suited for this role, since the data is collected across the country using standardized definitions. Other state and regional databases can also be used.
The first useful perspective we found was the breakdown of mortality by surgical priority or urgency (Table 1). Elective and urgent cases, which together comprised about 90% of cases, showed a lower mortality than the state average; but the mortality for emergency cases was higher than the statewide average.
|
|
Table 3 shows results for the high-acuity patients, those emergency cases with an acute MI less than 6 hours old, with shock, or that were hemodynamically unstable. Mortality began to decrease in 1993, with our process of care changes. There were very few deaths in this group after 1992. As indicated by the surgical volume, this mortality decrease was not the result of turning away high-risk patients.
|
|
In summary, we learned a few important points. First, having a database is just the beginning; the real goal is to use it to obtain helpful information. Second, we must be experts and leaders in the use of our data. No one else is better qualified; and we can’t afford to give away the initiative. Third, database information can provide us with new perspectives and useful insights into our care. It can highlight both problems and opportunities that may not have been previously visible from a narrower perspective [5]. Fourth, everyone involved in a program must share responsibility for it; they need to work as a team. Fifth, we must optimize the care process as an overall system, especially where the boundaries and gaps lie between specialty areas.
Acknowledgments
I would like to thank all my associates in Albany for the privilege of using information based on their cases, and for the opportunity to present it here.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
