Methods of Presenting Data to Facilitate Continuous Improvement
After the results have been analyzed, the next step is to graphically display and disseminate them. There are many ways to present data. First, the data can be displayed using a frequency distribution such as a histogram (Fig 1-8) or using a scatter diagram (Fig 1-9). Are the data normally distributed (ie, distributed in a bell-shaped curve), or are they skewed? The answer to this question affects the selection of statistical tools and analyses, and it can provide important insights into potential underlying factors. Alternately, 2 distinct subgroups may be found in the data and need to be defined. For example, care in solo practices may differ from care in large single-specialty groups for a particular disease area.
The purpose of these data analyses is to identify variation in the factor of interest. Factors that are due to the way the system is established and that are inherent in its current state of operations are called common cause factors. To improve performance in this area, the organization will have to redesign and reengineer the system. For example, there may be a known rate of “unreliable” visual fields in glaucoma, despite the best training of technicians and screening of patients. In contrast, there are “special causes” of variation that are due to a specific, identifiable factor, often a specific provider or person. Rapid identification of “special cause” variance allows for quick correction of variation that exceeds normal rates. However, improving the performance of the overall system and reducing the common cause variation will improve care and affect the most patients. By “shifting the curve,” the organization can improve care for every patient, as opposed to just identifying the outlier providers and assisting in their rehabilitation.
Excerpted from BCSC 2020-2021 series: Section 1 - Update on General Medicine. For more information and to purchase the entire series, please visit https://www.aao.org/bcsc.