I have just recently completed reading Atul Gawande’s latest book, The Checklist Manifesto. I have read all three of his books and I find him to be one of the most gifted and outstanding authors around, with fantastic readability and thoughtfulness. I highly encourage giving any of his works a perusal.
Standardization in Radiology has come to the forefront in our field over the last several years, with more calls for a radiology lexicon.
The argument for report standardization is being taken more seriously, as there are more calls for uniform reporting of pathology, moves towards pay for performance, and further accountability as far as quality and measurability of radiology reporting, as we head towards the future.
Similar to BI-RADS, a standard lexicon would be a roadmap for amalgamating disparate reporting techniques and terminologies for radiologists to more easily compare and communicate results. Right now, our Tower of Babel fragmented methods are quite subpar. In a future where Google-like data analytics software will be the norm for indexing radiology results (teaching files, billing, literature searches), I can see no other alternative but to adopt such a method.
The key point I took from Dr. Gawande’s book was that checklists (a proxy for standardized radiology reporting tools) are powerful apparatuses for breaking down sophisticated tasks into manageable and automatable tasks. It is a key way to diminish our attentiveness away from the redundant tasks of imaging report generation. As sophisticated as imaging has become, with multiple comparison studies, additional complex MRI sequencing and CT protocols, and more and more findings being reported -- a uniform lexicon would help approach the radiologic problem of complexity.
As intelligent as physicians are, no one is infallible. If more of our minds would concentrate on the images, and less on the actual structuring the report, I believe errors in radiology would diminish, which of course would be a great boon for our patients.
How do you think radiology should be standardized?