Examining Radiology with Statistical Analysis
There has been a great deal of conversation in the media regarding our healthcare system lately. With that in mind I thought I would share an example brought to me by one of my colleagues serving the healthcare industry, where the hospital client used statistical analysis for process improvement in one area of radiology.
The Goal: Reducing the Number of STAT Portable Chest X-Rays
When first commissioned, the team was asked to reduce the number of “STAT” portable chest X-Rays. Portable chest X-Rays are widely used to assist in diagnoses while attempting to minimize the inconvenience to the admitted patient. The medical procedure is not complex but the fulfillment of the request requires significant use of man-hours that may otherwise be redirected. (While it would be nice to think that the word STAT was short for statistics, for the uninitiated, it is derived from the Latin word statim, which means immediately.) Therefore, a STAT portable chest X-Ray compounds the demand on resources, since the request hits the process with the same impact as a custom order. Specifically, the request demands resources and manpower that was not previously scheduled and/or anticipated. While every system makes provisions for some customization of this type, in this case the number of STAT portable chest X-Rays made up 48 percent of the total portable chest X-Ray‘s for this hospital, a volume that significantly handicapped the performance of the department.
A Six Sigma Team Investigates the Issue
A Six Sigma team was deployed to address the issue. The Six Sigma team talked with the stakeholders of this process and determined that the reason for the inordinate number of STAT requests was the slow turnaround time of the routine portable chest X-Ray. The stakeholders indicated that they would be happy with a turnaround time of 90 minutes from the time of the order to the preliminary verbal report from the radiologist. When the Six Sigma team measured the actual performance of portable chest X-Ray‘s, however, they discovered that the mean turnaround time was 1,281 minutes, or 21.3 hours, with a standard deviation of 1,081 minutes or 18.0 hours. It was clear why so many STATs were ordered. The descriptive statistics of the process capability are shown below. (Click on diagram to enlarge.)
The Six Sigma team immediately began the standard process evaluation: defining the process, identifying potential causal factors for the defects and collecting and analyzing data. After several months of hard work, the single most significant factor was the time taken between “hanging” the image and the creation of the preliminary report, contributing 91.6 percent of the process variation from the mean turnaround time (TAT), as shown in the following regression analysis output. (Click on diagram to enlarge.)
The Six Sigma Team’s Steps to Improve the Film-Reading Process
Now that the key area had been identified, the Six Sigma team set out to devise a series of process improvements to the film-reading process. The following histogram graphically represents the pilot performance, utilizing these process improvements, compared to the baseline. (Click on diagram to enlarge.)
The solid blue column is tracking the post-pilot performance while the dotted areas represent the pre-pilot (baseline) performance. It is important to note that the post-pilot performance was still not meeting the stakeholder expectations for portable chest X-Rays. Nevertheless, the post-pilot data were statistically different from the baseline performance, as shown in the chart and t-Test below. (Click on diagram to enlarge.)
Once tested and proven, process checks were put in place to assure consistent and sustained performance reflecting the process improvement. But were the needs of the stakeholders met? The following chart shows the performance of the process post-pilot. (Click on diagram to enlarge.)
The process improvement achieved was a welcome beginning to continued improvement. The performance was reduced from a mean of 21 hours to approximately eight hours, and the standard deviation improved from 18 hours to approximately three hours.
The Six Sigma team will use statistical analysis to further reduce and monitor variation until they reach or exceed the stakeholder’s specifications. In the meantime, the delays have been significantly reduced, allowing for earlier diagnosis, treatment and eventual discharge. Improving these cycle times clears hospital resources for treatment, improves patient care and reduces the waste and/or cost in caring for patients.
A Call for Process Improvement Within Healthcare
Process improvement is certainly something that our healthcare system needs if we expect to meet the needs of our population and the goals laid out by government. Now if we could only get someone to look at the insurers for efficiency...
Have Confidence in Your Statistical Analysis!: Learning How to Use Confidence Intervals
The Right Tool Matters!
Pocket Stats, Part 3: Sample Size for Pass-Fail Tests
Sales, Six Sigma and Statistics: How One Team Used Statistical Analysis to Solve a Sales Challenge and Make Money
Pocket Stats: Quick Significance Tests You Can Remember, Part 2
Pocket Stats: Quick Significance Tests You Can Remember
Getting Through This Economic Downturn With Style
Accelerate Your Journey to Improve Customer Satisfaction and Reduce Operational Cost
May 13, 2014
6th Annual Nordic Enterprise and Process Excellence Forum
7A Odenplan, Stockholm, Sweden
May 19- 21, 2014
How to Avoid the 5 Pitfalls of Process Optimization
May 20, 2014
Transform Your Process Excellence Training: Skill-building, Culture Change and Employee Engagement
April 2, 2014
The Secret of Building a Sustainable Process Transformation Program
March 18, 2014