SCHD lab personnel noticed for some time that the totals for Gonorrhea and Chlamydia (G&C) tests in the “Lab Tests Performed” report from the KIPHS data management system did not match. Since both tests were to be completed on the same specimen, the totals should have been the same in any month or year-end reports. The data needed to be accurate for monthly reporting, CLIA certification, and reports to the Kansas Department of Health and Environment.

 

The totals of Gonorrhea tests performed as seen in the “Lab Tests Performed” report in the KIPHS data management system will match the totals for Chlamydia tests in the same report for each month from January 2009 through August 2010, as will the following months through November 2010. 

 

The QI Project team consisted of lab, information technology, and data entry/encounter processing personnel.  These were input experts for one of the three possible areas where the cause might be located. They represented some of the stakeholders. A timeline was established using a Gantt chart, a schedule-monitoring method, and general team member assignments were agreed upon.

 

The total number of Gonorrhea tests performed did not match the total Chlamydia tests in 2009 or 2010. It was discovered that the problem was mainly in the early months of 2009. After the tests were run a second time after the end of the month, the 2010 totals matched.

 

The problem appeared to be rooted in three possible processes:
a. Lab testing and reporting,
b. Numbers generation and reporting in the KIPHS data system itself,
c. Generation of encounter forms and data entry by provider and fiscal associates.

 

The developers of KIPHS were contacted to learn how the numbers were generated for the G&C totals in the “Lab Tests Performed” report.  It was discovered that KIPHS data management system was pulling the data from two different sources – test request date and test completed date - which did not necessarily match.  KIPHS redesigned the software in September 2010 to pull the numbers from the correct source – the test request date (which matches the encounter date).

 

The “Lab Tests Performed” report for G&C tests was run for 2009 through August 2010 after the redesign was complete. It was also checked for matching totals monthly from September through November 2010 to make sure that the totals continued to match.

 

The “Lab Tests Performed” report had been run before the changes were made to the software. Immediately following the changes in September 2010, the report was run again for the same periods (Jan.-Dec. 2009 and Jan-Aug 2010). It was found that the changes made in KIPHS corrected the problem going back to January 2009. The test results now match in every month during these periods.

 

The original improvement theory was proven incorrect since the totals did not match in September after KIPHS was corrected. However, when the correction was made in the design of KIPHS, the totals of G&C tests matched in every month in which they had not previously matched. This indicates that an improvement was made to the system. This particular improvement was standardized by redesign of KIPHS.

 

However, the September report showed that G&C totals did not match by one test. The discrepancy in the September test report indicates:

 

A. Establish a method with KIPHS to correct lab data inside the lab module after data has already been entered.
B. Create an option for a more detailed “Lab Tests Performed” report to show which clients did not receive both tests.
C. Develop a system to cross-check between the encounter form and the lab test requisition form.
D. Analyze the provider/encounter/data entry system for quality improvement to make it a viable way to validate the lab totals.