Publishing good science is the closest most of us get to immortality

First, a story about the last time I cried: a few weeks ago, I began to wonder how Dr. Susan Braithwaite, one of my professors from my endocrine fellowship at the University of North Carolina, was doing. I had asked her for a letter of reference in January of this year, 2022, because I was applying for a part-time job back at my medical school alma mater, the University of Kansas School of Medicine here in Wichita.

I emailed her to ask for the letter, which seems kind of strange and unprofessional except for the fact that she was a furious, enthusiastic emailer; during hospital rounds, she would go from floor to floor, opening computers and buzzing through her inbox to make sure no patient or physician had contacted her with critical information in the last fifteen minutes. In her immediate emailed response to my request, she mentioned she was having some health issues, but that she hoped to get back to work soon. She graciously agreed to write my letter of reference. It was promptly delivered to the Promotion and Tenure Committee at KUSM-W.

I got the job, and I wrote her an e-mail to thank her for her help. I offered to send her a physical, snail-mailed thank-you note, but I didn't know her physical address, and I don't think she was interested in giving it to me. In my thank-you email, I mentioned that I had used, and continue to use, her basal-bolus insulin titration algorithm that she, with an alliterative panache, published as “Five Steps to Freedom” on New Year’s Day in 2005. If you’re curious, here are the five steps:

But I’m getting ahead of myself. I promised a story about crying, not a story about blood sugars. A few weeks ago, after using the Five Steps algorithm for what must have been my 1,000th patient, I wondered how her health was holding up. So, I sent her a quick email.

I did not get an immediate response.

After a few days, I got a gmail nudge telling me that I had sent the email to Dr. Braithwaite five days earlier and asking, did I want to try again? I felt a wet, furry sense of dread. Instead of sending another email, I instinctively typed “Susan Braithwaite obituary” into Google and closed my eyes as I hit “enter.” Then I saw this:

And that’s when I cried. I cried for the reasons most of us cry when we encounter loss. I mostly cried for my guilt. I hadn’t reached out to her in January because I wanted to see how she was doing; I knew she’d taken a new job at Florida State, and I wasn’t curious enough to simply see how that was going for her. Nope. I reached out to ask a favor. And she gave it to me, even though she had maybe five months to live. It wasn’t even her first time doing me a favor. Once upon a time, she invited me to publish a letter with her in JAMA Internal Medicine, something I was underqualified to do. But she dragged me along, anyway, just like she helped hoist me into a new job at KUSM-W.

I also cried for the loss of such a beautiful, mathematical mind so dedicated to making the world a better place. But, in processing her loss, I thought of how many people will continue to benefit from her brief presence on Earth. She seemed to know her time was short, from the first moment I met her in 2006. She wouldn't talk to drug reps, a practice that I admired and now follow myself. I don’t think she saw them as a threat to her decision-making, the way some of us do. And there was no Sunshine Act back then, so she wasn’t protecting her reputation. I suspect she simply thought they were wasting her time. I wish I could say that she had endless time to spend with patients. Actually, I don’t. She didn’t have endless time to spend with patients. She had the exact amount of time it took to get an accurate history, do the proper exam, give advice, prescribe medications, and document the visit. She didn’t suffer through the rituals of the exam room that many of us believe make us better doctors. She refused, even, to type into the electronic health record. Instead, she would dictate a quick note in front of the patient, nodding to the patient to make sure she was getting the details right. Somehow (I really don’t know), the notes ended up in the EHR. They were shorter, more concise, and more useful than almost anyone else’s.

Instead, she spent her time using her mathematical mind to examine dose responses to insulin at varying blood glucose levels and coming up with novel treatment algorithms for critically ill patients with diabetes. I'm guilty of showing her graphs of insulin-versus-blood glucose levels in presentations of my own and intentionally breezing over them to hide my own mathematical shortcomings. And, though I can’t prove it, I suspect that modern-day smart insulin pumps use similar mathematical logic to what she was able to casually draw up on conference room whiteboards. To get a taste of this, see her 2014 paper on “Multiplicative standard deviation for blood glucose.” Or a dozen others. Just enter “Braithwaite SS” into Pubmed. You’ll find that her dozens of publications are pragmatic and aimed, laser-like, at the immediate, demonstrable improvement of patient care.

Sigh… my point is, Susan Shapiro Braithwaite lives on. She did the hard work of inching the science of medical delivery ever so slightly forward. And, for those of us in the academic world, that’s as close as we can get to immortality.

Are We Witnessing the End of the Pharmacy Benefit Manager?

As much as we’d like to think that every player in the healthcare marketplace is an integral part of the team, contributing to the health and well-being of our employees and saving employers money, it’s becoming harder and harder to support the ongoing abuses of pharmacy benefit managers, or PBMs. I won’t rehash them here. You can find past opinions about them in other posts.

Two discoveries this week at KBGH make me wonder if we’re witnessing the end of the PBM model. First, in a (unfortunately paywalled) editorial in the New England Journal of Medicine, Leemore Dafny describes the business model of a new drug company called “Civica.” Civica was launched in 2018 by seven health care delivery systems and three philanthropies to address chronic shortages of generic drugs, like IV antibiotics and sedatives, that are essential to inpatient care. Civica has since grown to sell about 50 generic injectable drugs to more than 1,500 hospital members. Their newest venture is the production of three “biosimilar” insulins (aspart, lispro, and glargine, which you may know better by their brand names Humalog, Novolog, and Lantus) that make up about 80% of the insulin market.

Here’s the important part: Civica intends to make its insulins available to “all U.S. pharmacies at the same wholesale price and on identical terms.” Everybody gets the same price, and Civica won’t engage in the usual PBM “rebate” trickery. It will print wholesale prices and the manufacturer’s suggested retail price right on the packaging! That MSRP will represent more than an 80% reduction from prices we’re accustomed to paying through PBMs, who keep about 60% of our money as “rebates” and other chicanery. The MSRP for a pack of five insulin glargine (a.k.a. Lantus) pens will be $55. Dafny invites us to compare that to the usual list prices of $425 for Lantus, $404 for Semglee, and $148 for unbranded Semglee.

The second cloud on the horizon for the future of traditional PBMs is Mark Cuban’s online pharmacy CostPlusDrugs.com. Instead of eliminating the PBM, Cuban has simply made the pharmacy and the PBM one and the same, with fully transparent pricing.

I don’t think we can carve the gravestone for PBMs yet; they surely have new tricks up their sleeves, and I expect legal action to be part of their strategy to try to delay companies like Civica from acting too quickly. But the tradewinds against the old, predatory PBM model seem to be blowing harder and harder. “This plan is straightforward: sell a drug at cost and at the same price to all buyers,” says Dafny of Civica. If only other health services operated the same way.

Quick tip of the hat to Drs. Bob Badgett, MD, and Kevin Wissman, PharmD, at the University of Kansas School of Medicine-Wichita for alerting us to these.

As the Medical Director of the Kansas Business Group on Health, I’m sometimes asked to weigh in on hot topics that might affect employers or employees. This is a reprint of a blog post from KBGH.

63% of What You Pay for Insulin Pays for Something Other Than Insulin

A landmark study (paywall) landed on our doorsteps last week not with a thunderous crash, as it should have, but with a gentle thud. Investigators from UCLA’s School of Public Policy published a cross-sectional analysis of where money paid for insulin actually went in the years 2014-2018.

An accompanying editorial (also paywalled) aptly describes the methodology of this research paper as “daunting,” given the difficulty in getting information in such an opaque marketplace. The investigators had to compile data on 32 separate insulin products from The SSR Health Rx Brand Pricing Data Tool, a commercial pharmacy claims database, CMS records, individual state Medicaid and drug transparency reports, and public filings of participant companies, including not only drug manufacturers, but drug wholesalers, pharmacies, pharmacy benefit managers (PBMs), and insurance companies themselves. Their methodology was complex enough that I’m not sure I completely understand it, and I’m confident I can’t pass it on to you in less than 800 words. So I’m going to get down to the nitty-gritty. First, here is their helpful schematic on the flow of money in the drug pipeline, which I found to be as good as about any explanation I’ve come across:

JAMA Health Forum

You can see that the key breakdowns in this complicated flow of goods and services are at points (2) and (7), where kickbacks from the manufacturer to the PBM (also known as “rebates”) are paid to move a drug up the formulary and where the PBM subsequently passes a share of those rebates on to the benefit plan.

Between 2014 and 2018, the average list price of the 32 insulin products rose 40.1% (inflation over the same period was 5.96%). So the list price of insulin rose eight times as fast as the list price of, say, dishwashers.

At the same time, though, the average net price paid to manufacturers fell by 33.0%. So, for example, Eli Lilly, the oldest insulin manufacturer in the United States, got $69.71 of every $100 spent on its insulin products in 2014. By 2018, they were getting only $46.73 out of every $100 spent.

Similarly, the share going to health plans (as those sneaky “rebates”) fell by 24.7%, and the share of expenditures retained by drug wholesalers increased by 74.7%, from $4.63 to $8.09, although the absolute costs were obviously low in comparison to others in the chain.

Reader, prepare your body for the coming rage (I recommend some light stretching and some John Tesh on the stereo). The share of insulin expenditures going to PBMs from 2014 to 2018 rose 154.6%, from $5.64 to $14.36 of every $100 spent. And the share retained by pharmacies increased by 228.8%, from $6.21 to $20.42.

If you can still read through all the red mist, let me reiterate this: Let’s say that you and your diabetic employee together pay ~$115 to insurance company X in December. Insurance Company X will then take its allowed Medical Loss Ratio of ~$15 and spend the remaining $100 on insulin. Of that $100, the breakdown of who gets what is below:

• Manufacturer: $46.73

• Wholesaler: $8.09

• Pharmacy: $20.42

• PBM: $14.36

• Health insurer: $10.40

Here’s the same information in graphical form, which is helpful to see the change over time:

JAMA Health Forum

Understand that about a quarter of people on insulin–your employees–report routinely skipping doses because of insulin’s tremendous expense. While those people are very likely shortening their lives because they can’t afford their drugs, PBMs, pharmacies, and to lesser extent wholesalers, are taking ever-greater shares of the money that you and your employees pay for the insulin, all while stiffing the manufacturer. I don’t mean to let the manufacturers off the hook; the pharmaceutical industry is the most profitable industry in America by a healthy margin. But we need to recognize that every participant in the distribution system for insulin, from the manufacturer to the PBM to the pharmacy, shares some of the blame for the astronomical increase in price that we’ve experienced over the last decade or more.

Outside of breaking out the pitchforks and torches, what can be done about this problem? A lot, we believe. As we’ve outlined before, you can look at your PBM contract to make sure you have access to your data, that you are allowed to access underlying contracts of your PBM, and that you can get out of your contract in a reasonable amount of time. With that information in hand, you can challenge new clinical approvals for low-impact drugs added to your formulary. You can even buy generics directly from the manufacturer. And, if you’re a member of KBGH, you could immediately cut pharmaceutical costs even without uncomfortable conversations with your PBM by working with our Right Rx program.

As the Medical Director of the Kansas Business Group on Health, I’m sometimes asked to weigh in on hot topics that might affect employers or employees. This is a reprint of a blog post from KBGH.

The biosimilars are coming! (and that’s a good thing)

As the Medical Director of the Kansas Business Group on Health I’m sometimes asked to weigh in on hot topics that might affect employers or employees. This is a reprint of a blog post from KBGH:

Most drugs on the market are made the same way you made new compounds in chemistry lab in high school or college: by putting molecules or elements together in a solution and adjusting the temperature, pressure, pH, and other inputs to encourage the elements to combine in a specific way. So if an enterprising high-schooler were to discover a way to produce pure, uncontaminated aspirin in her garage, we could be confident that it would have the same effect as aspirin produced by Bayer.

And generics—unbranded versions of name-brand drugs—really do work to drive down the cost of medications. A 2017 study in the New England Journal of Medicine found that for a drug with a single manufacturer, generic and brand-name prices were roughly the same. But for a drug with three manufacturers, the generic price was only 60% of the brand-name price. So it’s no surprise that 90 percent of drugs used in America are generic.

But newer drugs—and some old ones like insulin—are not synthesized from scratch. Because of their complexity, we have to trick bacteria into making them—as with insulins—or we have to isolate them from living organisms where they naturally occur. We call such large-molecule drugs “biologics,” and instead of calling copies “generics,” we call copies of these drugs “biosimilars.” (As a side note, the first bacteria-produced “recombinant” human insulin ever to be given to a human was injected right here in Wichita by Dr. Richard Guthrie.)

When we make a biosimilar drug to match a biologic drug, the new drug is not necessarily atom-for-atom identical to the brand-name drug. The FDA allows the new drug to be slightly different as long as “no clinically meaningful differences” are noted between the original drug and the biosimilar drug in terms of safety, purity, or potency. This obviously requires more study than what would be required for our garage-produced aspirin. This means a lot of work, much of it in human subjects. Because of this difference, biologics have long operated under a different set of rules than “small-molecule” drugs. Whereas old-fashioned drugs are granted a 20 year patent from the date of application, after which it is relatively straightforward for another company to start producing a generic version, biologics are subject to a kind of natural immunity to generic competition that’s made worse by bad behavior on the part of the reference drugs’ manufacturers, misinformation campaigns, and close-but-not-quite-similar clinical outcomes.

The result has been that, while biosimilar drugs are common in western Europe, their use has been very limited in America. As such, Americans have long assumed that we paid far more for biologic drugs than our peer countries have.

A new study proves this out. Pharmacists and physicians from the University of Pittsburgh looked at the price of the only four biosimilars on the market by December 2018 (white blood cell-producing filgrastim [Neupogen®] and pegfilgrastim [Neulasta®], immunosuppressive drug infliximab [Remicade®], and insulin glargine [Lantus®]) over time. Their findings were striking. From 2007 through 2018 the cost of each of these medications went up by ~5-14% per year. At the time of introduction of a biosimilar (or two years ahead of time in the cases of glargine and infliximab) the cost of the medications immediately plummeted: −7.7% for filgrastim, −7.4% for pegfilgrastim, −13.6% for infliximab, and −23.5% for glargine.

The timing of this information is good. The dam on biosimilars is breaking. Dozens of biosimilar drugs have been approved in the last two years. So as you work with your pharmacy benefit manager to design your drug benefit, make sure that biosimilars are covered for your most expensive biologic agents. This may be harder than you think; abusive contracting practices are some of the obstacles biosimilars face in getting into wide use. However, the benefit to your bottom line will be substantial if you’re able to successfully integrate biosimilars into your pharmacy benefit.

Links to health for Wednesday, August 9, 2017: Laennec and his stethoscope, rich people love endurance sports, the health of elite athletes, and mythical drug expiration dates

After Rene Laennec invented the stethoscope, he encouraged other doctors to make their own.

This is reminiscent of the way Banting and Best handled insulin: 

"...the first “sale” of insulin was for just $3 (Canadian), not for a vial, but the very intellectual property of the drug itself. Frederick Banting, Charles Best, and James Collip, the team that first discovered and refined insulin therapy, agreed to receive $1 each in exchange for giving their patent rights to the Board of Governors of the University of Toronto in 1923.

This was a bold move, and not without controversy at the time, according to an article posted by the University of Toronto Centre for Innovation Law and Policy. At the turn of the 20th century, it was considered beneath scientists and universities to patent medical innovation at all. However, Banting and company believed that they needed to patent their formulation of insulin to stop pharmaceutical companies from rushing to patent an inferior, less potent form of the drug. The university immediately gave pharmaceutical companies the right to manufacture insulin, royalty-free, and to improve upon the formulation and patent any subsequent improvements."

If only we saw such responsibility and modesty from drug manufacturers now.

But back to Laennec: it is strange that we continue to venerate the stethoscope 200 years after its invention, especially when we have rapidly evolving, already far superior equipment in the form of handheld ultrasound. As the folks at 99 percent invisible note, docs no longer routinely carry reflex hammers or pinwheels, but they still overwhelmingly carry stethoscopes. I do not. You can't auscultate an adrenal tumor, you know what I'm saying? Bedside ultrasounds are the future, period. They'll just reach full market penetration sooner in the US than in developing countries.

Why do rich people love endurance sports?

Brad Stulberg pontificates on why endurance sports are so popular among high-income people. 

"...data collected in 2015 by USA Triathlon shows that the median income for triathletes is $126,000, with about 80 percent either working in white-collar jobs—professions such medicine, law, and accounting—or currently enrolled as students."

Once upon a time, and still for a shrinking older population, golf was the sport that signaled wealth. But now, with the fuzziness that surrounds the measurement of effectiveness in the knowledge economy, endurance sports give white-collar workers a simple, quantitative way to measure progress. You run/swim/bike/ski faster or slower than you did the last time, period. And the pain doesn't hurt (pun intended):

“By flooding the consciousness with gnawing unpleasantness, pain provides a temporary relief from the burdens of self-awareness,” write the researchers. “When leaving marks and wounds, pain helps consumers create the story of a fulfilled life. In a context of decreased physicality, [obstacle course races] play a major role in selling pain to the saturated selves of knowledge workers, who use pain as a way to simultaneously escape reflexivity and craft their life narrative.” 

Link from marginalrevolution.com

Related: Are elite athletes healthy?

Answer: it's complicated.

The myth of drug expiration dates

A pharmacist finds a cache of drugs, unopened, all 30-40 years past their expiration date. He has them tested by a toxicologist. The result? All of the drugs--all of them--are still useable.

Congressman Earl Blumenauer (and his sweet bow tie) wants to reform the Farm Bill

Link from foodpolitics.com

Links for July 27, 2017: lead levels and crime, CTE, happiness in bariatric surgery, and the Fda's cheese problem

Are high lead levels associated with high crime rates?

In a sample fraught with selection bias, 99.9% of autopsied NFL brains had chronic traumatic encephalopathy.

This bothers me less than I might have thought it would. NFL players are small in number, they have a job that they know is risky, and they're pretty well paid for the work. The fact that they have a now well-documented complication of their employment is only mildly interesting. Where this research needs to go now is into lower levels of play. If evidence begins to accumulate that college, high school, or younger players are at risk for CTE, then wholesale changes are going to be needed in football or the sport is in real danger. 

People with a BMI under 40 are happier and healthier after their surgeries.

I'm a big, big fan of bariatric surgery for people with morbid obesity. We simply have access to no other treatment that improves quality of life and prolongs life to the extent that surgery can. But now there's some evidence (albeit weak) that people should have surgery earlier, not later. In a study published in JAMA Surgery, investigators found that people who had surgery before their body mass index, or BMI, hit 40 were about 12 times as likely as people with higher BMIs to get to a non-obese BMI (that is, less than 30) after surgery. A couple things to consider here: first, surgeons want to operate, so we should acknowledge their bias in this study; of course they want people with lower BMIs to qualify for surgery. It opens up more patients for them and they get to operate on a less sick, less complicated population of patients. But second, weight isn't the only thing we worry about post-surgery, so why does this matter? Because people who got to a BMI of less than 30 were much more likely to be able to stop medications for cholesterol, diabetes (especially insulin), and blood pressure. They were also much more likely to see a remission in their sleep apnea. Finally, the people who got to a lower BMI were just happier, with a satisfaction rate of 92.8% vs 78.0% for folks who didn't get to a non-obese BMI. The study can be tossed into the growing pile of studies showing that gastric banding sucks. Patients who had any other surgery--sleeve gastrectomy, gastric bypass, or duodenal switch--were 8 times, 21 times, and 82 times more likely, respectively, to get to a BMI less than 30 compared with those who got gastric banding. And that doesn't even take into account the disturbingly high complication rate of gastric banding. Take-home points? If you're considering bariatric surgery, get it sooner rather than later, and don't get a gastric band. 

The USDA is all about cheese right now.

I have mixed feelings on this. On the one hand, it seems a little absurd that we so misallocate food subsidies that dairy farmers are literally pouring milk into holes in the ground:

Farmers poured out almost 50 million gallons of unsold milk last year—actually poured it out, into holes in the ground—according to U.S. Department of Agriculture data. In an August 2016 letter, the National Milk Producers Federation begged the USDA for a $150 million bailout.

On the other hand, given the growing evidence that excess meat intake is shortening American lives, diverting some of that desire for meat toward cheese, yogurt, and other dairy products isn't the worst thing in the world. 

June 2, 2017 links to skim

Michael Bliss, author of The Discovery of Insulin, has died. RIP. I read his book while I was an endocrine fellow, overlapping, ironically, with a trip to Toronto for the Endocrine Society conference. His work was accessible, non-academic, and revealing.

“Family-based weight loss therapy sessions worked just as well whether children attended or not, as long as their parents did,” researchers found in a “two-arm trial” that included “150 overweight or obese children ages 8 to 12 and their parents.” Parents, your work with your children matters a lot. 

Whole-body vibration may be as effective as exercise in mice. Color me skeptical. I first heard of vibration for metabolic disease in a presentation by an astronaut in college in the nineties. And, in spite of the plethora of machines available claiming to vibrate you to better health, the science just doesn't seem to have advanced that much. I'll skip my vibration sessions for the time being.