Normal Curve - An Old Idea with Timeless Application

We all learned about the normal bell shaped curve, aka bell curve, sometime in our education. So what? During stat class, I learned about it to  be used for calculating t and z score to find p values. I just now appreciated it's timeless value.

One of the most important ways to improve ourselves is to know where we stand with respect to our colleagues. For example, I'm a radiology resident and it's important for me to know where my knowledge base is with respect to my colleagues at my level. Some of the measures that can be used include knowledge, accuracy and speed of interpretation. Our department recently acquired a new software called Montage, which analyzes imaging reports over the past three years. The software extracts and collates information which can then be analyzed and interpreted. One of the valuable functions that Montage provides is information about where we stand with respect to her colleagues in some of these quantitative measures, such as volume of exams we have interpreted. In addition, there are additional information you can extract like case volume over certain period of time, types of imaging studies done, etc. etc. I think it's really important for us to know these functionality because our department will be using quantitative parameters to measure our performance against the departmental benchmark. More importantly, we can use information to measure our own performance against of those around us and help us understand where we need to be in comparison to those we work with.
More importantly, most of us, if not all of us have missed something significant in imaging findings over the course of our careers. The missed finding may have been significant impact on patient's care and outcome. This happens to all of us at some point in our career and is independent of our sub-specialty. It is not uncommon for us to really take it hard on ourselves and not be able to recover from it as quickly as we could. Many of us interpret these missives as a human deficiency, rather than as a human error. Having recently attended a breast imaging conference, I learned about how we can apply the normal curve to help ourselves and improve our situation. In breast imaging, all breast imagers are accountable for their reports. They have follow up pathology findings of their imaging interpretation. As required by federal regulation, performance measures are closely followed for every radiologist and enter performance is compared to the national average. Through this effort, the federal regulators have developed a normal curve of expected misses.  Having this information helps us understand where we are what respect to our colleagues at the national level. This was certainly an "aha" moment for me.

The national health quality initiative that is occurring is understanding the value of normal curve. The idea of improving quality, a.k.a. reducing iatrogenic injuries, will be achieved by reducing the range of the normal curve and thus narrowing the standard deviation (see figure). This is in contrast to what has previously been done, quality assurance, which was thought to be a static process of evaluating mistakes and placing blame. Placing blame is not effective because it does not improve overall quality per se and the second, this old process lack continual effort to improve current processes. A cycle of Plan, Do, Study and Act (PDSA) is one example of quality improvement advocated by the American College of Radiology. In this process, we first recognize the problem, then, develop a plan to address the issue with a goal in mind, followed by a period of testing the new plan. Lastly, we study the results and repeat the cycle over and over again until we reach our goal. Personally, the act of studying our results and knowing the outcome of the actions is the most important part of this process.  Feedback is probably the most important part of our learning and making us more effective at our outcomes. An example of how this process is used to improve quality in imaging is demonstrated by the breast imaging subspecialty. At the Los Angeles Radiological Society Breast Imaging conference, I learned from one of the speakers that the most important predictor of a successful and effective breast imager is not their image volume, it's not where they practice, but rather, whether they got  feedback from the pathologists about their reads or not! Breast imaging has really set the standard for quality improvement for field of radiology.

The idea of normal curve can be applied in all aspects of our lives and can be used as an objective measure to understand where we are and where we need to be to be the best we can be.