It’s July, and that means a new generation starting radiology call. I’m not sure I’ve ever done a listicle or top ten, so here are fifteen.
- Look at the priors. For CTs of the spine, that may be CTs of the chest/abdomen/pelvis, PET scans, or *gasp* even radiographs.
- Look at all reformats available to you. On a CT head, for example, that means looking at the midline sagittal on every CT head (especially the sella, clivus, and cerebellar tonsils) as well as clearing the vertex on every coronal.
- Become a master of manipulation. If your PACS has the ability to generate multiplanar reformats or MIPS, don’t just rely on what the tech sends as a dedicated series. Your goal is to make the findings, and you should be facile with the software enough to adjust the images to help you make efficient confident decisions, such as adjusting the axials to deal with spinal curvature or tweaking images to align the anatomy to the usual planes when a patient is tilted in the scanner. MPRs are your tool to fight against confusing cases of volume averaging.
- Your reports are a reflection of you. I don’t know if your program has standard templates or if those templates have pre-filled verbiage or just blank fields. There is nothing I’ve seen radiologists bicker about more than the “right” way to dictate. What is clear is that you should seriously try to avoid errors, which include dictation/transcription errors as well as leaving in false standard verbiage. We are all fallible, and Powerscribe is a tool. Do whatever it takes to have as close to error-free reports as humanely possible.
- Seriously, please proofread that impression. Especially for mistakingly missing words like no.
- Templates and macros are powerful, useful, and easily abused tools, just like dot phrases and copy-forward in Epic. I am all for using every tool you have, but you need to use them in a way that comports with your psychology and doesn’t make you cut corners or include inadvertently incorrect information.
- Dictate efficiently. If you are saying the same thing over and over again, it should be a macro. If you use PowerScribe, you can highlight that magical text and say “macro that” to create a new macro. (On a related note, “macro” is a shorter trigger word than “Powerscribe”.)
- More words ≠ more caring/thoughtful. As Mark Twain famously said, “I didn’t have time to write a short letter, so I wrote a long one instead.” It’s easier to word vomit than to dictate thoughtfully, but no one wants to read a long (or disorganized) report. Thorough is good, but verbose doesn’t mean thorough. It usually means unfiltered stream of consciousness. The more you write, the less they read.
- Never forget why you’re working. The purpose of the radiology report is to create the right frame of mind for the reader. Our job is to translate context/pretest probability (history/indication) and images (findings) into a summary that guides management (impression).
- Address the clinical question. This is especially true in the impression. If your template for CTAs was designed for stroke cases and says some variation of “No stenosis,” that impression would be inappropriate for a trauma case looking for vascular injury.
- Include a real history. Yes, there are cases where an autogenerated indication from the EMR is appropriate, but there are many more where that history is either insufficient or frankly misleading/untrue. You need to check the EMR on every case for the real history. Then, including a few words of that history is both the right thing to do and also very helpful for the attending who is overreading you.
- Radiologists practice Bayesian statistics every day. This is to say: context matters. A subtle questionable finding that would perfectly explain the clinical situation or be more likely given the history should be given more psychological weight in your decision-making process than if it would be completely irrelevant to the presentation. For example, a sorta dense basilar artery is a very different finding in someone acutely locked-in than somebody with a bad episode of a chronic headache.
- Work on your tired moves. We can’t all make Herculean calls at 4 am. When you’re exhausted and depleted, you rely on the skills you’ve overtrained to not require exceptional effort. For radiologists, this boils down to your search pattern. You need to not just have well-developed search patterns but also to have sets of knee-jerk associations and mental checklists of findings to confirm/exclude in different scenarios to prevent satisfaction of search (e.g whenever you see mastoid opacification in a trauma case, you will make sure to look carefully for a temporal bone fracture).
- Everyone is a person. The patients, the clinicians, the technologists, and any other faceless person you talk to on the phone. It’s easy to feel distanced and disrespected sitting in your institution’s dungeon. But even you will feel better after a hard night’s work when you’re a good version of yourself and not just someone sighing loudly and picking fights with strangers.
- Music modulates the mood.