Objective: It's important to determine reliable reference intervals for all the test parameters of the clinical laboratory, but it's usually hard to obtain useful and necessary clinical data. Although the procedure of indirect determination of reference intervals using stored data is not defined in any guide, it was showed that the reference intervals could be estimated by indirect method in lots of studies. We aimed to constitute a model of using stored data in verifying the reference intervals of prothrombin time (PT) and activated partial thromboplastin time (APTT) with exclusion of certain clinics in this study.

Materials and Methods: Results of PT (N=11,363) and APTT (N=7,034) test requests of ambulatory patients aged 15 to 80 made from outpatient clinics of surgical departments before surgical interventions in 2008 were retrieved from the electronic medical record. Certain clinics were excluded. Thromborel S and Actin (Dade Behring, Germany) were used for the tests which we done on the Sysmex CA-1500. We used Bhattacharya and Hoffmann methods as a retrospective quality control procedure for each study week. When we compared the sensivities of these methods, the results were similar.

Results: We determined the weeks where there was disagreement between the results of internal quality control and patients' tests and excluded the results of these weeks in estimation. When we compared the sensivities of these methods (Bhattacharya and Hoffmann), the results were similar.The indirect reference intervals estimated by using the stored data with the exclusion of certain clinics and the retrospective evaluation of the internal quality conrol results were overlapping with the data stated in the insert.

Conclusion: The model which we constituted using stored data, may be particularly suitable for confirming reference intervals of the laboratory and evaluation of the results of the presenting population.