Thread by @TheWonkologist, "Are Financial Payments From the Pharmaceutical Industry Associated With Physician Prescribing?"Our latest [...]

"Are Financial Payments From the Pharmaceutical Industry Associated With Physician Prescribing?"
Our latest on physician-industry COI is out today in @AnnalsofIM!

Not to bury the lede, the answer is "yes."
Thread to follow.
https://www.acpjournals.org/doi/10.7326/M20-5665
1/

First things first, huge thanks to everyone on the research team: Niti Trivedi, Susan Chimonas, Sara Tabatabai, Johanna Goldberg @mskcc_library, Renee Gennarelli, @ldiaz1, @DKorenstein. This was an enormous amount of work that we are all proud of.
2/

The idea for this study grew out of this old Twitter thread of mine. I *thought* I had read everything on this topic. But...was I subconsciously cherrypicking? Was the story more complex? Only one way to find out.
3/ https://twitter.com/TheWonkologist/status/1060367563459321866?s=20

https://twitter.com/TheWonkologist/status/1060367563459321866?s=20

So we conducted a systematic review.
All included studies assessed the association between financial payments from the drug industry (meals, consulting, gifts, speaker fees) and physician prescribing practices.
4/

The studies we found spanned all branches of medicine. Most used Open Payments data to measure industry payments. Most studied Medicare prescribing. Most were from 2016 or later.

But you know what they ALL had in common?
5/

THEIR RESULTS!!
Every last study that we included found that industry $$$ is associated with physician prescribing. Specifically, prescribing more of the drug made by the company that they got the payments from.
6/

(Well, almost. Some studies had several co-equal primary analyses, and sometimes some of these analyses had null findings. But in all studies 50% or more of the primary analyses found a positive association)
7/

Here's the overall distribution of the individual analyses across all 36 studies, in terms of whether the association was positive or null, grouped by the risk of bias of the study.
8/

So the first thing I always hear when I talk about these studies is "correlation is not causation." It seems to be a prevalent view that it is entirely coincidental that the doctors who get the most money from companies also prescribe the most of the same companies' drugs.
9/

But you know what? Sometimes correlation IS causation. How do know? Because we found these studies:
Agha et al
Orlowski et al
@citizeness et al
10/

But what's that you say? Physicians need to work with pharma to drive innovation? Great! Let us do it. Let's work with pharma on developing new drugs.
But, just to be crystal clear, that is NOT what these studies are measuring.
11/

These studies are measuring the $2.4 billion in personal, NON-RESEARCH payments to physicians each year, for POST MARKET products. These are all payments for drugs already in widespread use.

The innovation has already happened. These payments are for persuasion.
12/

Other highlights of our review include several studies that found evidence of how pharma payments make us worse doctors.
13/

Here's a study from myself and @DusetzinaS @AaronWinn2 showing how payments from @Novartis push physicians from imatinib to the more-expensive and more-toxic nilotinib when treating CML
https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/10.1001/jamainternmed.2018.0776
14/

Pharmaceutical Industry Payments and Oncologists’ Selection of Targeted Cancer Therapies in...

This study examines the association between oncologists’ receipt of payments from pharmaceutical manufacturers and drug selection in 2 situations where there are multiple treatment options.

https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/10.1001/jamainternmed.2018.0776

Here is a slew of studies, including work by @joefigs2 @SAgrawalMD @cpgYALE @DrScottHadland, and by @ki_endoepi using strong causal-inference methods, showing that industry payments increase opioid prescribing:
https://pubmed.ncbi.nlm.nih.gov/30566426/
https://pubmed.ncbi.nlm.nih.gov/30667135/
15/

Pharmaceutical payments to physicians may increase prescribing for opioids - PubMed

In the United States, physicians who receive direct payments from providers for opioid prescribing tend to prescribe substantially larger quantities, particularly for hydrocodone and oxycodone.

https://pubmed.ncbi.nlm.nih.gov/30566426/

...oops, too many to fit in one tweet:
https://pubmed.ncbi.nlm.nih.gov/32350126/
https://pubmed.ncbi.nlm.nih.gov/29799955/
https://pubmed.ncbi.nlm.nih.gov/31011967/
16/

This paper by @manvi_rx and @DrHollyHolmes &co shows how industry payments get us to prescribe really useless branded combo pills like saxagliptin+metformin and (seriously?) ibuprofen+famotidine
https://pubmed.ncbi.nlm.nih.gov/29609611/
17/

So to summarize
1. All available studies show an association between pharma payments an prescribing
2. This association is causative
3. This results in worse prescribing practices
18/

Historically, physicians have been in favor of these payments.
https://pubmed.ncbi.nlm.nih.gov/14557219/
https://pubmed.ncbi.nlm.nih.gov/20566978/

Which is understandable - who doesn't like free lunch?

19/

But we now know that accepting industry payments is bad for our patients. Knowing this, we need to stop dragging our feet and change our behavior.
Now.

20/

In 2009, the Institute of Medicine encouraged us to end industry COI through self-regulation before forced to do so through external regulation. A decade later, little has changed.

Perhaps it's time for that external regulation?

21/

You can follow @TheWonkologist.

Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: