To the Editor.—
Quality tools and information technologies can be used to manage metadata regarding the use of laboratory tests. Benchmarking and the laboratory information management system can be used to search and manage data from large laboratory patient data bases.
The Appropriate Utilization of Laboratory Tests group (REDCONLAB) was created in 2010 to build a network of shared knowledge among clinical laboratories with the purpose of providing health services of the highest quality in the context of management excellence. This REDCONLAB study compared the requests for laboratory tests from primary care providers in different health care departments in Spain. In these studies, a final report was sent to participating laboratories, comparing their own results to those from their counterparts.1 Interpractice variability and test request inappropriateness was significant.
In the REDCONLAB studies, each participating laboratory provided their production statistics (number of tests performed) obtained from their laboratory information management system for a certain year, for the entire population covered. With this data, appropriateness indicators (test requests per 1000 inhabitants or ratios of related test requests) were calculated by a single coordination center. In the 2012 edition (76 laboratories, 17 679 195 inhabitants), the final report suggested some recommendations to improve the request appropriateness of aspartate aminotransferase (AST), antigliadin antibody (AGA), antithyroglobulin antibody (TgAb), total bilirubin (TBIL), γ-glutamyltransferase (GGT), iron, free thyroxine (FT4), erythrocyte sedimentation rate (ESR), and urea tests, with a list of test-reduction strategies that worked in primary care facilities.2 In the 2014 edition (110 laboratories, 27 798 262 inhabitants), every participant laboratory was asked to report what strategies were implemented to deal with the aforementioned inappropriateness, the implementation date, and which tests were they aiming at.
In this letter, we discuss how those strategies affected the request of the aforementioned laboratory tests. The performance indicators to study the effects of such interventions used the following ratios of the requests for related tests: AST to alanine transaminase (AST/ALT); AGA to antitissue transglutaminase antibodies (AGA/anti-tTG), TgAb to antithyroid peroxidase antibodies (TgAb/TPOAb), TBIL/ALT, GGT/ALT, iron to ferritin, FT4 to thyroid-stimulating hormone (FT4/TSH), ESR to C-reactive protein (ESR/CRP), and urea to creatinine. Those indicators from the 2014 REDCONLAB edition were compared in health care departments in which the strategy was implemented before January 1, 2014 (strategy group), and those in which the strategy was not implemented before December 31, 2014 (nonstrategy group), via the Mann-Whitney U test. A 2-sided P ≤ .05 rule was used as the criterion for rejecting the null hypothesis of no difference. Analysis was performed using SPSS for Windows, Version 22.0 (SPSS Inc, Chicago, Illinois).
The Table displays the number of laboratories that implemented improvement strategies for each particular test, as well as the results for all the different performance indicators. All the indicator results improved in the strategy group and were statistically significant for AST/ALT, TBIL/ALT, iron/ferritin, FT4/TSH, ESR/CRP, and urea/creatinine. The latter ratio was more than 6 times lower in the health care departments whose laboratories implemented improvement strategies than it was for those laboratories that did not.
Abbreviations: AGA, antigliadin antibodies; ALT, alanine transaminase; TgAb, anti-thyroglobulin antibodies; TPOAb, antithyroid peroxidase antibodies; anti-tTG, antitissue transglutaminase antibodies; AST, aspartate aminotransferase; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; FT4, free thyroxin; GGT, γ-glutamyltransferase; IQR, interquartile range; TBIL, total bilirubin; TSH, thyroid-stimulating hormone.
P ≤ .05.
This reflects how results obtained from comparative studies in large populations can be used to increase the effectiveness of a health care system or organization, in this case, to optimize the request of laboratory tests whose use is not recommended as first-line liver,3 thyroid,4 renal function, anemia,5 or inflammatory markers.6 The scope of our study results could be the financial effect these types of strategies could generate. Unfortunately, only a few of the participating laboratories implemented improvement strategies, despite the recommendation, suggesting that more education and communication is needed, so that benefits can be generalized to the entire health care system.
Metadata obtained from current, routine, available technologies through collaborative initiatives are useful for improving the appropriateness of laboratory test requests.
The authors have no relevant financial interest in the products or companies described in this article.