Goldman Sachs and the Optics of Drug Discovery

Goldman Sachs analyst Salveen Richter, channeling the obvious in a note to clients (excerpted by CNBC):

The potential to deliver ‘one shot cures’ is one of the most attractive aspects of gene therapy, genetically-engineered cell therapy and gene editing. However, such treatments offer a very different outlook with regard to recurring revenue versus chronic therapies. While this proposition carries tremendous value for patients and society, it could represent a challenge for genome medicine developers looking for sustained cash flow.

Ew, go on (emphasis mine):

GILD is a case in point, where the success of its hepatitis C franchise has gradually exhausted the available pool of treatable patients. In the case of infectious diseases such as hepatitis C, curing existing patients also decreases the number of carriers able to transmit the virus to new patients, thus the incident pool also declines … Where an incident pool remains stable (eg, in cancer) the potential for a cure poses less risk to the sustainability of a franchise.

Yes, franchise. Long term profits depend on the riskiness of a cure.

I’m not going to begrudge a private company their desire to make money. The possibility of windfall profits are the main reason why private companies are willing to invest in uncertain and risky biomedical research. That said, when the long tail of a too-good cure only makes tens of billions in profit, it should be hard for even a staunch capitalist to be sad.

This attitude is part of what drives fringe antivaxxers and other patients away from evil “big pharma” and the medical doctors who understand the actual practice of medicine and into the arms of pseudoscience. For my part, I don’t think any company should feel bad if they develop an HIV vaccine so effective it eradicates the disease and relieves the suffering of millions, even if it eventually results in downstream profit loss due to the loss of the chronic antiviral therapy market.

We badly need and will always need public and government research support—for many reasons—but one is because the optics of the patient as a customer mindset are so toxic.

Stop Free-Dictating

There are many institutions/practices with well-defined “normal” templates for all types studies, which help provide a reasonable approximation of a house style. A clinician (or the next radiologist) has a reasonable chance of knowing where to find the information in the report. The reader can see something in the impression and quickly find the longer description in the body of the report for more information.

Templates can be brief skeletal outlines or include more thorough components containing pertinent negative verbiage. A section for the Kidneys could say “Normal” or it could say, “No parenchymal lesions. No calculi. No hydronephrosis.” Some groups have diagnosis-specific templates that build off a generic foundation to better address specific concerns like renal mass characterization or appendicitis.

Either way, some form of templating is critical to creating a readable report. After all, radiology for better or worse is a field where the report is the primary product, and creating reports that are concise, organized, and readable should be a goal.

Some institutions and practices do not have these baseline templates. There are (often but not always older) attendings who seem to not only practice but respect the freewheeling old school transcriptionist style of reporting. A resident who doesn’t “need” a template is to be prized and congratulated.

This isn’t 100% wrong either. It’s a useful ability in the sense that it’s important to be able to summarize findings in cohesive English. It’s largely the same skill as the casemanship skills used during hot-seat conferences that the recent Core exam generation of residents have largely lost, and so I can appreciate this perspective. However, at least from a reporting perspective, this is wrong in the 21st century.


The purpose of the radiology report

The first attending I ever worked in radiology was a neuroradiologist who posed a semi-rhetorical question on my first day. He used to ask:

What is the purpose of the radiology report?

The answer, he argued, was to create the right frame of mind in the reader.

I think this view is exactly right.

Defined in a narrow sense, this means that the reader should come away with the impression that you intend for them to have. If something is bad and scary, that should be clear. If something is of no consequence, that should also be clear. Items in the impression are there because we want those impressed on the minds of our readers, not just because we saw them.

With increasing patient access to radiology reports, we now have a second audience. While doing away with all medical and radiological jargon is probably misguided and unnecessary, we need to at least be cognizant of how our reports might read to a layperson (or non-specialist, for that matter). If we can be more clear and more direct, we have a greater chance of communicating effectively to all involved parties.

Templates make reports more organized, scannable, and readable. Not even debatable.

But while the primary intent of “frame of mind”-creation may relate to the significant radiological findings, it’s also about creating the right frame of mind about you, the radiologist. Thorough, thoughtful, organized, conscientious? Or rushed, disorganized, careless, apathetic?

There may be some perks of blinding readers with science and drowning readers in long-winded descriptions of even benign and irrelevant incidental findings. At least you won’t look lazy! But for the less verbose among us, we can show we care by creating reports that reflect our systematic approach and clear writing style. Templating is critical to creating digestible reports.

Lastly, as quality metrics rise in importance and resource utilization re-enters the arena as a responsibility of the radiologist, we also need our reports to be readable and indexable by computers. The easier our reports are to parse, the easier we can extract meaningful data about our findings, link these up with patient data from the EMR, and draw high-powered conclusions about patient impact, outcomes, and (of special importance to me) the utility of certain exams in specific clinical contexts.


Dictation software is a tool, not a recorder

If you’re a resident somewhere and your institution doesn’t have power normals to frame-out your reports, make some. If you find yourself saying the exact same things over and over again every single day, then you’re doing it wrong. It should either part of the template or an auto-text macro. If nothing else, it will reduce your rate of transcription errors.

No one needs to reinvent the wheel on every case!

The ABR’s new Online Longitudinal Assessment (OLA)

It was super duper gratifying to receive my first OLA email from the ABR this past month. OLA (Online Longitudinal Assessment) is the ABR’s new longitudinal MOC (Maintenance of Certification) process, where diplomates take 52 questions every year instead of a big test every decade.

I took the Certifying Exam in October and received my passing result in November, so the month-long break prior to needing to “maintain” my brand new certification from the ABR feels just about right. Yes, a thousand folks need to maintain a piece of paper they haven’t actually received in the mail yet. I can appreciate why folks fresh off their q10-year MOC victory are irritated at needing to immediately participate in more MOC. Promises are being broken left and right. But, hey, money.

Adding insult to injury, as a neuroradiologist, I still have to sit for the exorbitantly expensive ($3,270) neuroradiology subspecialty exam this October. Which means that I need to maintain my first certification in between getting my second.

The final irritant in this system of paying $340/year (forever) is that the ABR, which is a nonprofit sitting on a war chest of ~$48 million, didn’t apply for (i.e. pay for) ACCME accreditation, so the hours spent doing OLA questions don’t count as official CME.


The Actual OLA Experience

The current OLA paradigm is that 2 questions are released every week (104 a year) and “expire” after 28 days. So while you can log in and batch around 8 questions a month, you won’t be able to do it less often without losing some expired questions. Since you only need 52 questions and can do around 8 a month, you could actually get away with doing it almost bimonthly.

I took my first 8 questions this week and got them all right. They were straightforward, reasonable, and relevant to practice (at least in neuroradiology). My initial impression is that OLA questions are more like what the Core exam should be. You get between 1-3 minutes per question, the website was pretty slick (at least on a desktop), and I did all 8 in around 5 minutes. Can’t complain there. This is clearly a better system and more logical way to fulfill the spirit of MOC than taking an exam full of (even more) irrelevant material every decade.

You get to choose your practice profile and thus what types of questions you receive. I originally chose general diagnostic radiology and neuroradiology, but out of my first 8 questions, 7 were neuro and only 1 ended up being general, and the general question concerned GI fluoroscopy, which I detest, so I switched to 100% neuro. Maybe it’ll help with the subspecialty exam.


Things the ABR should improve:

  • Mobile experience. I’ve heard complaints about display issues on phones. You only get a minute for most questions, so it needs to work.
  • Lower the price. At the current rates, this is far more expensive than any commercial qbank. And that’s what this is. The ABR makes a lot of profit for a non-profit.
  • Increase the question-life. Why do questions expire after 28 days? So arbitrary. Let the radiologists hold themselves accountable. How about 90?
  • Get official CME accreditation. This feels like apathy and laziness. I know it’s not straightforward or cheap to be a CME-granting organization with the ACCME. But again, this is an expensive process, but it would be far more reasonable if it counted for CME.

And finally, how about you let everyone take the certifying and subspecialty exams using the OLA software instead of flying out to Chicago to waste their time?

Step 1 keeps you safe from the dangers of fun

If students were to devote more time to activities that make them less prepared to provide quality care, such as binge-watching the most recent Netflix series or compulsively updating their Instagram account, this could negatively impact residency performance and ultimately patient safety.

That’s Peter Katsufrakis, MD, MBA, president and CEO of the National Board of Medical Examiners (NBME) and Humayun Chaudhry, DO, MS, president and CEO of the Federation of State Medical Boards, responding in Academic Medicine to a student-written article concerning how Step-prep has consumed medical education that advocated for a pass/fail Step 1.

There was a backlash, and they tried to backpedal on this comment (emphasis mine):

During the editing process of our manuscript, we added a statement about excessive use of Netflix and Instagram which was unfair and inappropriate. As leaders of the USMLE, we believe that students, medical educators, and the public deserve our respect. Our statement was inconsistent with that belief, and we are deeply sorry.

Yeah, right. Make no mistake, their glib response to actual student concerns is exactly what they meant to say. Humor is often the dull dagger of truth, seemingly softer and more palatable than direct honest communication but ultimately more damaging.

However, the disrespect is by far the lesser evil here. Students and residents are rarely respected on an intellectual level by administrators. Their perspectives are viewed as myopic and ill-informed. The real issue here is dismissal.

Students have valid concerns. Residents have valid concerns. Trainee complaints are often dismissed by their superiors as the whining of a coddled generation (whether decades ago or today), and then those graduates go on to perpetuate both the toxic culture and broken system it engenders.

The biggest problem in medical education is the uncanny ability of doctors to pay-it-forward instead of being agents of change.