Posts in: science

Godwin’s Law and the Rise of Hyperbole on MedTwitter

Saurabh Jha:

“Today’s Galileos fight over one or two vaccine doses in teenagers, whether the risk of vaccine-induced myocarditis is 1/1000 or 1/10, 000. Nothing encapsulates our pettiness more completely than our probability wars.”

Echoes of Chris Hitchens here.

A brief chronology of my employment

  • 1994: Fifth grade; I am charged with editing the school newspaper. There is an Intel 386 PC at home that is about to be upgraded to a 486 and do something more than run Lands of Lore.
  • 1996: Seventh grade; I typeset a book of poems1. The school newspaper becomes the school magazine — in layout only; the publishing schedule remains haphazard — as I upgrade from Word 6.0 to QuarkXPress
  • 2000: High school starts again after a freshman year interrupted by NATO bombing. I make the town library’s official website. It is a php hack job laid out in tables instead of the newfangled and to me unknown CSS; it still wins an award.
  • 2002-2008: Med school; I typeset a book here and there and occasionally help out with the library website.
  • 2009: Teaching assistant, Institute for histology and embryology, Belgrade School of Medicine.
  • 2010: Resident, Internal medicine, JHU/Sinai, Baltimore MD.
  • 2013: Chief resident, Internal medicine, as above; I understand the benefits of not being invited to a meeting.
  • 2014: Clinical fellow, hematology/oncology, National Cancer Institute, Bethesda MD.
  • 2016: As above, but also Chief fellow ex tempore for the joint NCI/NHLBI fellowship; my hatred of poorly-run meetings intensifies.
  • 2017: Staff clinician, later to be renamed Assistant research physician, Clinical Trials Team, Lymphoid Malignancies Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda MD; the 1994 me marvels at the word salad trailing the title.
  • 2021: Chief Medical Officer, Cartesian Therapeutics.

  1. Someone else’s, to be clear. ↩︎

Blogroll

I, for one, am glad that blogs are making a comeback. Here are a few I’ve been reading for at least a few months, many of them for years, some for decades.

Applied philosophers

The only true philosophers of our time.

The new scientists

People without major academic credentials who have interesting ideas about science.

The old scientists

People with major academic credentials and interesting ideas, something to teach, or both.

The ludites

People against modernity of one sort or another.

People doing their own thing

Unclassifiable but exhilarating.

Apple enthusiasts

Some tips, a few tricks, many opinions.

Finance-adjacent

Economists and investors, for the most part.

Journalist-cum-substackers

Former or current journalists who now earn some or all of their living by writing newsletters via Substack, which is slowly reinventing blogs (in the sense of reinventing the wheel, not actually making them better and in fact in many was making them much worse).

Company blogs

For when I really want to know when the next update is coming.

Clearing the PDF log jam

There is a crisis in medicine, but not the one you think: And not only in medicine, of course.

Reading primary literature is superior to press releases and tweets — it sounds so obvious, but not many physicians act on it. There no prizes to be won for not just following the KOLs[^kol], Key Opinion Leaders, the influencers of medicine before influencer became a real noun. Note that unlike the influencers of social media KOLs don’t use the #sponsored hashtag, though there is a hashtag equivalent. nor do you save any time. Quite the opposite: instead of a promoted tweet about the me-too drug de jour falling into your lap, you need to find a way to identify what’s worth your time reading, and also find time to actually read it — not a small achievement, as highlighted by the above tweets.

But then what? Sure, there is profit at the end of the rainbow in the form of useful knowledge, but merely reading a PDF may not result in any knowledge at all, let alone knowledge you can use. Or, as the [underpants gnomes][gnomes] would put it:

Underpants gnome meme with the text Phase 1: Read PDFs, Phase 2: ?, Phase 3: Profit

I too had a backlog of unread PDFs once, spent so much time organizing files and folders, using this and that program to store the metadata, NB: if you write any sort of scholarly texts you will still need a reference manager, no matter what system of organizing PDFs themselves you choose. I recommend Zotero, lest your institution has a requirement for Endnote (which must have quite a salesforce, to so thoroughly insert their buggy, laggy, slob of a program into every academic crevice). trying out plain paper, a Kindle, an iPad or two, thinking it is how I was reading them that mattered and oh if only I could find the perfect setting, under the shade of an old oak tree perhaps, with some peace and quiet, a pen in one hand and a cup of coffe in the other, well, then the unread pile would melt away and all would be good in the world.

But reading is easy, if what you read is useful, entertaining, or both. For most people without visual impairments or dyslexia, the log jam is at Step 2. We don’t want to read our pile of PDFs because, in most post-GME circumstances, there isn’t a clear goal to reading them (lest you have superhuman memory). The clear exception here being board exam and MOC prep, where the goal is obvious and the sources of information are all spelled out. This is particularly true early on in your carreer, when you have nothing to hang your hat on mentally, and few connections to make between what you are reading and what you already know. Sure, you don’t need to keep track of the articles you’ve read if the only reason for reading is to pan them on Twitter. You do, however, want to summarize what you’ve read and save it for future use, be it in a lecture, article, grant proposal or a blog post. So if and when you find a fairly obscure but potentially important fact about this or that cellular pathway in a supplemental figure from a CNS-adjacent journal, and you memorize the fact for later use, and then a year or so later you do use it to make a figure for the background section of a clinical trial protocol, well, what you do not want happening in that case is to spend hours of your life trying to retrace your steps and figure out the original source when a fellow Yes, this has happened to me. We do have good fellows. asks where you got the data.

I wouldn’t be admitting to all that if I didn’t think I’ve found a solution. A few years ago, I replaced the unsustainable routine of just-in-time literature reviews for whatever I needed done with a robust knowledge management system — a GTD® © David Allen Co. 2001. It is a good system though. for ideas, if you will. It got to a point where I can read at least one article every day and skim a few more, get the useful information out and into my app of choice The app of choice before DEVONthink was Roam, which is a web service and a marvelous one at that, but unfortunately not much into encryption, privacy, and other things people dealing with confidential information like to have in the tools they use., and have all the information I need to write an editorial like this in a morning or two.

As with most of the things I do it is too personal and Rube Goldberg-y to be of use to anyone else, but it started with a forum post and a book, and if you’d like to turn your plate full of PDFs into something more usable may I recommend that you start with one or both of those and see how it goes. Could it be any worse than what you’re doing now?

What I believe that most people probably don’t (no data behind this, just the armchair)

The world in general, and the US in particular, is spending too much on goal-directed, targeted biomedical research while undervaluing both applied and theoretical physics. Picture Leonardo da Vinci drawing helicopters: that’s the modern-day cancer researcher. The universal cure for cancer — and there should be one, if humanity survives long enough to create it — will not come from an NIH grant. If grants are involved at all, it will be something initially funded by the National Science Foundation. The current system of funding (government, non-profit, biotech, you name it) is broken, and if you account for the opportunity cost it is a complete disaster. Each of these statements deserves at least a paragraph, but I am saving my carpal tunnels for a manuscript, an LOI, and a couple of protocols (oh, the irony).

In the meantime, a few things physician-scientists should do for the overall good:

  • find causes and create better prevention strategies, because a look at the SEER database will tell you that it’s not just bad luck;
  • eliminate barriers for administration of known curative therapies world-wide (do we really want to leave this to politicians and economists?);
  • ensure rapid and honest evaluation of the many new treatments, procedures, and diagnostic/prognostic methods coming out of the biomedical behemoth.

How beneficial any of this would be for one’s career is a different question altogether, but let’s not get into incentives because RSI. I am also very open to opposing opinions, since my being wrong would make my life easier.

Storytellers

Last week I shared a brief reflection on a tiny aspect of my commute. Please check it out it if you haven’t already, it is a quick read.

Wasn’t that nice? It started by introducing some old concepts in a new light—you knew about trains before, and maybe even knew there was a MARC Penn that line goes from Baltimore to DC, but probably didn’t know the specific trains and their timetables. Then it gave you a coherent explanation of a phenomenon you hadn’t known about before. This first caused slight, but not unpleasant, cognitive strain while you were figuring out what I writing about, followed by the small pleasure of an ah-hah moment once the pieces clicked.

It was a brain massage, if you will. It was also complete bull.

Not that anything I wrote was wrong, as far as I know, but I didn’t give many arguments for it being right, either. There were no ridership statistics or arrival times to back up my claims. And even if there were—I didn’t give any alternative hypotheses to explain the situation, nor reasons why those would be less likely than my own explanation. When you think about it, it was more of a brain Twinkie than a massage—all empty calories, with a fleeting feeling of fullness.

Welcome to 99.99999% of the written word, and to anything ever spoken out loud.

We like stories. They need to make a threshold amount of sense (this is why societies universally ostracize schizophrenics). They should contain an element of surprise (it is not that the 7:07 train would come later than the 7:23—twists like that do not surprise anyone any more—it is that it comes in much earlier because people think it wouldn’t). And they get bonus points if—as my last parenthetical implied—they paint the others as stupid or incompetent. There are many more checkboxes; more of them checked, the better the story.

Most professions are based on storytelling. Doctors tell different stories to their patients, each other, and themselves—as do most other scientists, to a different degree. Lawyers tell stories to their clients to make them believe they will craft good ones for the judge, jury, and the opposing side. Ask a marketer what makes a good commercial (spoiler: story).

Being a coal miner doesn’t involve telling stories. No one wants to be a coal miner.

Our minds prefer a good story over a true one, and will have us believe it more, too. However, the more boxes you see checked, the more suspicious you should be that someone manipulated the tale to make it more pleasurable, ergo memorable, ergo believable.

(So, if what you’ve just read made sense…)

If you are looking for an objective truth—or getting as close to it as possible—any medium that involves audio/visual queues will be an impediment. Sights and sounds stir up emotions, and emotions prime us to believe or not to believe. Pay attention to the background music in a documentary, or how the desk of that shifty lawyer they’re interviewing is a complete mess.

TV news is, of course, a joke—this is why comedy shows are becoming the most popular delivery form.

Written word has its own way of deceiving—anecdotes, incomplete data, misquotes, lazy references—all to make a better narrative. Just read anything by Malcolm Gladwell. And look at the time it takes to get to the bottom of just one tiny factoid in that story of the iron content in spinach. Finding truth is exhausting and exasperating, and people whose job it is to find it (hello, accountants) are way less fun than those who make stuff up. Mark Twain said it best:

A lie can travel half way around the world while the truth is putting on its shoes.

Misquoted? Most likely. Or is Huff Post wrong? It wouldn’t be the first time.

There is nothing in this post that bigger and better minds than my own haven’t written about already. But that’s a boatload of pages! Not many people have the time, discipline, and interest to read all that—and even if they did, they would keep making the same mistakes over again, as shown in several studies described in those same books (yes, yes, all studies are flawed; one windmill at a time, please). These things are hard-wired, and for a good reason—evolution doesn’t care for objective truths.

Or maybe it does. I don’t know, I’ve just made it up.

Statistics resources for clinicians

Another week, another Quora question.

What is an online resource for learning statistics needed for clinicians explained in a language that could be understood by doctors?

There are many biostatistics courses available on Coursera. Living in Baltimore, I’m biased towards JHU’s offerings. “Case-Based Introduction to Biostatistics” by Dr. Scott Zeger is a good one. If you prefer text to video, here are three good resources:

If I had to pick one, it would be Dr. Brush’s book. He is a cardiologist writing for other physicians in a language they can understand. Also, Richard Lehman recommended it, which is more than good enough for me.

What is the evidence for that?

This has become the mantra of every medical student, intern, and resident wanting to appear smart on rounds and conferences, of every attending intent on shooting down a team member’s suggestion. Five, ten years ago it might have have signaled genuine interest. Now it means, usually, “I don’t know anything about the subject, but I’m still calling you out on (what I think is) your BS. Here, look at me! I am evidence-based!”

No, nobody has posed me that question in quite a while, and I don’t remember ever asking it in any context. Although I understand asking questions means showing interest, I’ve always preferred looking things up myself. This would make me appear either very smart or very dumb, depending on whatever subcontious impression I made on the person in the first few minutes of us meeting. Try to use the halo effect to your advantage. But, honestly, except for a few very well-known examples listed in this excellent post, you can find “evidence” in the medical literature to back up any claim. Off-the-cuff conversations during lectures and rounds are not the best place to dissect them, especially when one side has seniority.

Research during residency

Of the three pillars of medicine, research is the most ellusive. Unless you are in an MD/PhD program—not an option for most Europeans—you will have other priorities in medical school. And unless your residency program has a built-in research year, the way most surgical residencies do, you will either be way too busy in a university or a large community program to do any research, or have plenty of free time in a lower volume community hospital that doesn’t have many research opportunities.

When I interviewed residents-to-be last year, my first thought on seeing a non-PhD applicant having 18 publications on his or her CV wasn’t “Wow, she is a research machine, we gotta have her”, but rather doubt that anyone could be that productive during medical school. More points subtracted for thinking the interviewers would be so gullible.

I graduated six years ago, far enough not to be able to give advice on how to do research as a medical student. The hows and whys are institution-specific, so anything I wrote would have to be in Serbian anyway. Residencies, though, are similar enough to each other that I do have some words of advice for new residents wanting to do Research! in a community hospital, university-affiliated or not.

  • Patient care trumps research. Unless you have already worked as an attending in another country before coming to the US for residency, don’t waltz in to your PDs office on day one asking about research opportunities. Prove yourself on the field first, then six months later, when you’re comfortable managing DVT prophylaxis, septic shock, and what not, start asking questions.

  • Get your own idea? Common wisdom says it is better to come up with your own question and start your own projects, since you will be more invested in the outcome. Well, yes, sort of. Unless it is a quick-and-dirty chart review you can do over a two-week vacation—and even then there are IRB hoops you’d need to jump through to get anything done—you will get your inexperienced self into the murky world of project management. Many brilliant ideas have died on the field of required signatures, ambiguous data points, and impossible-to-coordinate meetings. Which is why this next advice is important.

  • Find good mentors. Surrounding yourself with a few good people is orders of magnitude better than having many good ideas. Research topics come and go, as does our interest in different fields of medicine (yesterday’s apoptosis is today’s epigenetics is tomorrow’s something or other). It is unlikely that the research your started in residency will continue onward into fellowship, but the knowledge, skills, and general wisdom you pick up from your mentors should serve you well into your career. NB: don’t wait for someone to be “assigned” to you—although that’s what many residency programs will do. Seek out people who match your character and who would be able to give you advice in at least three fields: patient care, research methodology, and research topics. This can be one person, or five. And if you find an awesome mentor who just isn’t doing any research right then, you can always write a review.

  • Is it Science! or quality improvement? ACGME is big on Quality! and Patient Safety! this year. Programs take notice. If you can present your interest as a quality improvement project rather than small-s-science, consider doing it. Not only does showing interest in quality improvement look good on a CV, your institution might have special funds for resident QI projects. A dedicated QI mentor is also a good resource, if you want a carreer as a Sith lor—erm, hospital administrator.

Interest in research goes from I just want something on my CV so I could get a fellowship to When I grow up, I’ll have my own lab, but this applies to most people in most circumstances.