The Name Game: An Improved Search Algorithm for Finding CKD Literature
What is in a name? The field of nephrology has grown accustomed to name changes. It seems we are never satisfied with only one name. Is it the Renal Division or the Division of Nephrology? How about Renal Week, or what we now call Kidney Week? What used to be acute renal failure is now acute kidney injury. The list goes on and on. This makes conducting research or searching for published literature confusing and sometimes difficult. In a recent article published in AJKD, Iansavichus et al from the London Health Sciences Centre in Ontario, Canada, describe the creation and testing of two high-performance information search filters to identify studies related to chronic kidney disease (CKD). CKD has a broad range of terms used to describe this clinical “entity”. These include chronic renal/kidney -disease, -failure, -insufficiency, -dysfunction, decreased renal/kidney function, pre-uremic, pre-dialysis, and pre-ESRD/ESKD. The same problem arises in the basic science literature. For instance CKD could be termed diabetic nephropathy, glomerulosclerosis, or kidney/renal fibrosis depending on the study involved. Iansavichus et al first developed and then validated and tested two search algorithms for CKD: one with great sensitivity thereby casting a broad net and another with great specificity and thus more narrow in its scope. These can be accessed at the McMaster University website.
The algorithms were created by using search terms identified after reviewing clinical practice guidelines, textbooks, and the MeSH (Medical Subject Headings) thesaurus and typically associated with diminished kidney function. In order to validate these algorithms, relevant articles were established by reviewing 39 separate journals. In order to test how well these filters worked, they were tested with clinical queries to 5 different nephrologists (see Table 2). An example question was “what are the effects of statins on change in kidney function and urinary protein excretion?” The use of both algorithms helped improve the accuracy of finding relevant articles by about 2 to 3 fold. This is a fantastic tool and the website is easy to use. It would be great if they developed these for use on a smartphone as well (currently this is not supported). The same group has also developed search algorithms for other areas of nephrology such as kidney transplantation, glomerular disease, dialysis, and AKI. You can go to their website and use any of these wonderful tools to further refine your queries on any topic related to nephrology.
Dr. Matt Sparks
AJKD Blog Advisory Board
To view the article abstract or full-text (subscription required), please visit AJKD.org.
Leave a Reply