Statistical process control & the operating table
Statistical Process Control (SPC) & the operating table: How SPC tools improve the outcomes of cardiac surgery
EXAM QUESTION: What is SPC and why would you give a decision maker/manager a control chart?
In order to achieve quality standards numerous companies use statistical process control (SPC) within their business. SPC is a technique used to control processes and determines whether or not material is accepted/rejected based on a degree of variability. For the most part, SPC is used as a corrective action when a product or service is being produced but the technique can also be harnessed for other purposes. In the healthcare field, SPC has been used as a tool for research and healthcare improvement. Data taken from healthcare improvement efforts can be effectively illustrated through control charts, which are an easier communicating medium for decision makers to understand in comparison to pages and pages of statistics (Benneyan, Lloyd, & Plsek, 2003). As an example, let’s look at cardiac surgery.
Even with the development of cardiopulmonary bypass techniques; cardiac surgery, like any other major surgery, poses a risk to the patient. To give you an example, repairs of congenital heart defects have an estimated 4-6% mortality rate (Stark J, 2000). So why does SPC fit into this picture and how can it improve the outcomes of cardiac surgery?
Doctors and nurses rely on numerous monitors to track a patient’s heart rate, oxygen and blood sugar levels and it is these monitors that alert them when a patient falls below or above their expected level. In the same way, SPC has the ability to monitor a patient’s health and in the case of cardiac surgery, facilitate near “real time” performance through the measurement of key clinical indicators such as blood product/reoperation, major acute post-procedural complications and length of stay/readmission (Smith, Garlick, Gardner, Brighouse, Foster, & Rivers, 2013). Monitoring of these outcomes allows for early detection/intervention of patient morbidities.
To appropriately illustrate this information a population-based study was performed using simulation and statistical process control. This took form in the monitoring of risk factors from a clinical database using empirical data on cardiac surgery. Cardiac surgery already uses advanced risk models such as euroSCORE and the aim of the study is to illustrate the use SPC methods to graphically show what would occur if you upcode (game) data. By gaming a risk you are able to see if a single factor can greatly change the evaluation of data.
Data was obtained from the adult national cardiac surgery database of the Netherlands Association of Thoracic Surgery. All patients who underwent cardiac surgery between January 1, 2007, and December 31, 2009, in all 16 cardiothoracic surgery centers in the Netherlands were included. Data on 46 883 consecutive cardiac surgery interventions were extracted (Siregar, et al., 2013).
Expected risk factor frequencies were based on 2007 and 2008 data and are illustrated in the below chart.
To summarize, figure 1 shows a single risk factor for women in one hospital. The top chart shows a mean frequency in 2007-2008 of 29% along with standard deviations. Through this chart expected frequency is calculated for 2009 (black portion) and you can see that each month falls within the limits. At no time does this risk go beyond the limits.
Now in order to gain insight into the outcome of patient risk upcoding (gaming) is performed. Intentional gaming of risk factors the upcoding (gaming) in random patients was simulated and detected in 100% of the simulations. These risk factor frequencies were based on 2007 and 2008 data.
After gaming was simulated the chart was constructed again. Monthly frequency rates of 18 risk factors in 2009 can be seen below:
The gaming lead to a clear rise in the mean euroSCORE as illustrated above. Not only has the mean frequency risen to 79% but in reviewing the chart you can see multiple outliers (red dots). Risk factors are a definitive issue during these months and are alarmingly high, out of control and outside of variability due to assignable causes. For
So at the end of the day, what does this all mean?
By conducting this controlled simulation we are able to see the use these SPC tools in field of cardiac surgery. The graphical display gives a means to monitor changes in risk factors. Through the observation of overall expected risk and also single risk factors, doctors are able to more accurately predict not only a patient’s euroSCORE but also determine a patients probable post op morbidities; allowing for early detection/intervention. Additionally, by reviewing a larger data source (10 year timeframe, multiple risk factors) doctors will have a better outlook in understanding the fluctuations from the frequency.
From a management standpoint, the ability to view data this way gives a decision maker key insights into the data in an understandable form.
Benneyan, J., Lloyd, R., & Plsek, P. (2003, December). Statistical process control as a tool for research and healthcare improvement. Quality & Safety in Health Care, pp. 458-464.
Siregar, S., Roes, K., van Straten, A., Bots, M., van der Graaf, Y., van Herwerden, L., et al. (2013, January). Statistical Methods to Monitor Risk Factors in a Clinical Database: Example of a National Cardiac Surgery Registry. Circulation: Cardiovascular Quality and Outcomes, pp. 110-118.
Smith, I. R., Garlick, B., Gardner, M. A., Brighouse, R. D., Foster, K. A., & Rivers, J. T. (2013, February). Use of Graphical Statistical Process Control Tools to Monitor and Improve Outcomes in Cardiac Surgery. Heart, Lung and Circulation, pp. 92-99.
Stark J, G. S. (2000, March 18). Mortality rates after surgery for congenital heart defects in children and surgeons’ performance. Lancet, pp. 1004-1007.
What is EuroSCORE? (n.d.). Retrieved 03 04, 2013, from euroSCORE.org – The official website of the euroSCORE cardiac surgery scoring system: http://www.euroscore.org/what_is_euroscore.htm