How to Categorize Incident Reports to Fuel Quality Improvement
A Model-Based Approach for Process and Behavior Incidents

Peggy A. Ahlin, BS, MT(ASCP) Director of Quality and Compliance ARUP Laboratories Salt Lake City, Utah	Bonnie Messinger Quality Manager ARUP Laboratories Salt Lake City, Utah

What is a practical approach to categorizing incident reports so that incident reporting fuels quality improvement?

As with any task, it is important first to define the goal. What information do you wish to capture? And more important, what do you plan to do with it?

The obvious intent is to facilitate process improvement and/or promote behavior change. Some incidents can be traced to poor process design, while others are the result of human fallibility. In general, it makes sense to categorize process improvement changes by areas of focus, such as process control systems; however, behavior change categories should center on interventions.

Table 1 provides an example of a useful model for categorizing behavioral incidents. This model is helpful because it provides guidance for the most appropriate interventions. For example, lab managers would respond to employee judgment errors related to the lack of rules by creating the necessary rules and training to those rules. Judgment errors related to reckless or intentional behavior would be reduced or eliminated by disciplining or terminating the employee, or finding a better fit for the employee in a position with less risk.

Incidents involving reckless violations of rules, 
like destruction of computers in the laboratory, 
usually mean that the employee is not a good fit for the workplace.

Likely candidates for the process improvement categories are those used in your organization’s quality systems. This approach allows you to focus on improvement strategies that match the quality system you want to improve.

For example, according to the Clinical Laboratory and Standards Institute’s Quality Essentials model, mislabeled specimens would fall within the process control system. Adding tiers of categories such as pre-analytic (mislabeled specimen), analytic (testing of a mislabeled specimen), and post-analytic (reporting a result inconsistent with a historical value) could also help refine the focus of your response. Other possibilities for categories are the major steps in your organizational flow chart or value stream map.

The number of tiers you choose will determine the granularity of your process review. One caution: with each higher level of granularity, the number of categorization choices will increase exponentially. If there are too many choices, users will have difficulty finding the most descriptive sub-category. If this happens, the data is less valuable for focusing your quality improvement efforts.

In summary, the best approach to categorizing incident reports enables users to accurately categorize both process- and behavior-based incidents, significantly helping your quality improvement efforts.