Exception Focused Review - background - Tip284
OBJECTIVE or GOAL
Exception Focused Review: essential background information
ENVIRONMENT
- Empower
- Empower Tip of the Week #284
PROCEDURE
Exception Focused Review is a technique used to minimize the time required for data review when using Empower. It provides increased confidence that potential data integrity issues will not be missed during review, by automating the process and eliminating the need to know or reference SOPs for specific values. The first tip in the series provides essential background information. Subsequent tips will describe how to create the Custom Fields and View Filters to carry out Exception Focused Review.
STEP 1
Leveraging the ISPE GAMP Records and Data Integrity Guide, data review should not be a "rubber stamp" on the data provided but a series of assessments, starting with:
- Do I understand how this data was created, and can I trust that these values truly represent the sample under test?
- Are the values within limits?
- Is it safe to release product based on these values (Figure 1)?
STEP 2
Dr. Raymond Panko at the University of Hawaii spent time looking at human errors (reference: http://panko.shidler.hawaii.edu/HumanErr/Index.htm). In his study, Dr. Panko found that people are good at making errors but not good at finding them (Figure 2).
STEP 3
Nevertheless, people are good at some things, such as spotting patterns and following a trail of events. Software is very reliable at assessing large amounts of data and automatically checking results against defined limits. Combining people and software is a powerful approach; the software can screen results for suspect data and people can investigate flagged results for Data Integrity issues (Figure 3).
STEP 4
The PIC/S definition of exception reporting is a validated search tool that identifies and documents predetermined "abnormal" data or actions for further attention or investigation by the data reviewer (Figure 4).
STEP 5
The Oracle relational database builds relationships between the results and all the pertinent pieces of information that help determine how that result was created (Figure 5).
STEP 6
View Filters are tools that query the database and display information in such a way that you can easily see any anomalies in the results (Figure 6).
STEP 7
Using Boolean logic in Custom Fields can indicate whether a series of ID numbers is consistent (Figure 7).
Empower can quickly scan results using a combination of View Filters and Custom Fields (using Boolean or Enumerated logic) to identify anomalies in our results. Reviewers can then investigate those flagged results.
ADDITIONAL INFORMATION
This can be done with either the Pro or QuickStart interface.