Imagine you’re shopping in the supermarket cookie aisle and a gorgeous, charming salesperson for Nabisco comes up to you and says, “We’ve noticed that you used to buy Oreos, but lately, you’ve switched to Nutter Butters instead. We’d really like to get you back to Oreos. May I offer you these free samples?”
I don’t know about you, but I’d feel a pretty deep sense of offensive intrusion. Yet that is pretty much how it works with prescription drugs: salespeople have immensely detailed information on the prescriptions written by each doctor, and they can use it to make their marketing as specific and effective as possible. Only in the drug industry, it’s much more concerning than the cookie industry.
Prof. Kevin Outterson, co-director of the health law program at Boston University School of Law, points out: “If it’s Oreos, it’s only a snack and I’m buying them for my family. What we worry about for physicians is that they’re making an important medical decision for somebody else. We’re trusting the physician to make the right decision, without inappropriate influence from drug companies.”
In the latest New England Journal of Medicine, online today, Kevin writes here about Vermont’s recent attempt to fix what we could call “the oreo problem” in drug sales. Vermont passed a law that barred pharmacies from selling doctors’ prescribing data to data miners and drug companies — unless the doctors themselves opted in. (Names of patients are protected by HIPAA, the federal medical privacy law, but names of doctors are not.)
The Supreme Court recently shot down that Vermont law, so drug companies remain free to buy information that helps them market their drugs better to each doctor. Kevin’s article analyzes that decision and looks ahead at its implications. We asked him to explain.
Q: Why is this issue important enough to merit inclusion in a top medical journal?
The medical profession has been interested for a while in how pharmaceutical companies are marketing to doctors. They’ve devoted a lot of articles and commentary over the years to whether doctors are making correct clinical decisions or whether they’re being inappropriately influenced by advertising and marketing and promotion.
Being able to identify prescribing patterns is very powerful to the companies. If they send in a drug detailer with a certain tactic, or if they meet with the doctor at a convention, or if they hire the doctor to give speeches, the company can track in real time, every day, the impact of all these activities. If a certain doctor is given free samples of a drug, they can then track over the next weeks or months whether the doctor has written new prescriptions for the drug. If the doctor doesn’t, they can go in and try to modify their behavior. They can ‘punish’ doctors who are not writing the scripts and reward those who do. If they know a doctor has recently switched and is prescribing more of a rival drug, they can go in and say something negative about the rival drug.
Data-mining raises questions as to whether physicians are being inappropriately influenced. That’s why medical societies and journals are concerned. They want to protect the integrity of the physician’s prescription process. They want doctors to be making decisions based on the clinical facts, with their patient, not based on the fact that the drug company knows too much about them.
Q: So could you sum up the Supreme Court’s reasoning in knocking down the Vermont law?
Vermont was too honest. Vermont said in the introduction to the legislation: We think doctors are being inappropriately influenced and our goal is to intervene in the ‘marketplace of ideas,’ and we’re going to do that by cutting off this flow of information from the drug companies and data-mining companies. Vermont inappropriately used a phrase (“marketplace of ideas”) that irritated the Supreme Court. It was an unnecessary self-inflicted wound.
If Vermont had framed this statute around medical privacy, they would have won 6-3, but since the case was about intervening in the marketplace of ideas, they lost 6-3.
Q: The Supreme Court cited the First Amendment in its ruling, but I don’t understand. I think of the First Amendment as defending political freedom of speech.
This is not Thomas Jefferson’s First Amendment. I think the founding fathers would be stunned to think corporate invasions of personal privacy are somehow protected by the First Amendment.
The justices assumed lots of people have this prescribing information, and what Vermont was doing was denying it specifically to drug marketers. Since the 1970s, the commercial speech cases in the Supreme Court have said that companies should be allowed to advertise and put information into the market as long as it’s ‘truthful and not misleading.’
One of the suggestions in my New England Journal of Medicine article is that the Court left a clear roadmap on how to fix the problem: Instead of saying the state is intervening in the marketplace of ideas, which automatically makes you lose five or six votes on the Supreme Court, take the Court’s cue that they would have decided it differently if it were a privacy case. Under HIPAA, the federal medical privacy law, there are already more than a dozen categories of protected health information, including patients’ names and full zip codes. Just make an additional category which is the name of the doctor on the prescription. If you do that, then it transforms the case into a privacy case, not a First Amendment case.
Q: In the article, you mention the American Medical Association also offers an option for doctors to keep their prescription information private.
The AMA makes tens of millions of dollars a year selling their data, called the Prescription Master File. It’s a significant percentage of the revenue of the AMA, some years more than $40 million a year in royalties.
The data-miners buy all the prescription records from the pharmacies and then they go over to the AMA and buy the physician master file with more than 600,000 doctors in the country, only one-third of whom are members of the AMA. The data-miners combine this information and sell it to drug companies. That’s how it ends up on the sales rep’s iPad, that Dr. Bob and Dr. Sally wrote three scripts yesterday; two were filled at Walgreens and one at CVS.
The AMA has had some criticism that they make so much of their annual income from data mining, to put it bluntly. So they’ve instituted a program that allows physicians to opt out of the data file. Just a couple of problems: One is that they don’t really publicize it, so while a number of doctors have opted out, it’s not a huge number. Two thirds of US doctors aren’t members of the AMA and may not be aware the AMA is doing it.
A second problem is that even if doctors agree, their patients haven’t consented to data-mining. Third, even if the doctor opts out, the AMA still sells the data to the data-miners, but with the promise that it only goes to the regional manager of marketing and not to the actual detailer. Finally, the opt-out expires after a couple of years and needs to be constantly renewed.
The AMA has a serious conflict of interest on this issue. The Vermont Medical Association voted unanimously to support the legislation. In their statute, Vermont gave doctors the opportunity to choose to be data-mined, and only a small number of physicians agreed. So it’s a tremendously different thing when you have an opt-in system.
Q; Is there anything we as patients can do?
Unfortunately, there’s no HIPAA right under current law to keep this data out of the hands of the data-miners. You can ask your physician if they’ve opted out through the AMA data restriction program. It will be interesting to see if your physician knows what you are talking about. More likely, they will say, ‘Huh?’
(Kevin also blogged about the New England Journal piece on The Incidental Economist here. )