Mining for Gold in Prescribing Records

Drug companies spent $5.0 billion on marketing prescription drugs directly to consumers last year and, as I have written in the past,  these ads have paid off. Studies have found that every dollar spent on DTC ads generates up to four dollars in additional sales of new drugs that are often only marginally better than far cheaper, generic versions or older drugs.

But as companies cut advertising budgets and introduce drugs that command smaller markets, there are signs that the days of $100 million-plus DTC ad campaigns might be over. A recent report by TNS Media, (a company that tracks media ads) found that DTC drug advertising was down 8% last year. Another survey of pharmaceutical industry marketers found that 58% of them plan to decrease DTC advertising this year, up from 28% in 2008.

One reason for the slowdown, according to the TNS report, is that companies are introducing fewer blockbuster drugs; the newest entries are approved for narrower uses with fewer potential patients and weaker sales projections. The report doesn’t mention that consumers also appear to be fed up with incessant advertising for erectile dysfunction, sleep problems and overactive bladder, and might be tuning out many of these mass-market missives. Drug companies also are facing keener oversight from policy makers who question the educational benefit of some DTC advertising and are examining their roll in increasing health care costs.

So where do drug companies go next to market their drugs? It turns out that many companies are refocusing their marketing efforts on an old, familiar target—physicians and their prescription pads.



 Pharmaceutical companies, of course, have always marketed directly to doctors—to the tune of more than $7.2  billion a year according to the Pew Prescription Project. That averages out to about $8,800 per doctor—although clearly some physicians (psychiatrists, for one) are targeted more heavily than others.

Much of that industry money is spent on detailing; the legions of sales reps who regularly visit physicians’ offices, pulling rolling bags stuffed full of free drug samples, journal articles and other giveaways. Drug companies also pay for doctors to travel to tropical islands and interesting cities to attend CME conferences, and they pay doctors to speak about the benefits of proprietary medications. They advertise in medical journals, fund research and send articles extolling the benefits of their drugs to physicians—including papers that recommend off-label uses. The euphemism commonly employed by the industry is that much of this pedaling is in the interest of “physician education.”

But, alas, public outcry—and the threat of federal legislation like the Physician Payment Sunshine Act that would require companies to report any payments or services provided to physicians valued at $100 or more—have caused companies to scale back on some of these time-worn methods. In January, for example, the industry trade group PhRMA voluntarily agreed to have member companies stop giving “branded” gifts (pens, pads, T-shirts, etc.) to doctors. They also promised to stop paying for physician meals and are requiring that money allocated for continuing medical education (CME) programs be handled by personnel who are not from sales and marketing departments.

To many, these voluntary efforts are hollow gestures; they do nothing to rein in industry-paid speaker fees, journal article handouts and other marketing ploys that are likely to have greater influence on doctor prescribing practices. And they also do nothing to address “data mining:” a growing—and controversial—industry practice in which companies like IMS Health and Verispan collect data on physician prescribing habits from pharmacy benefit managers and insurers and then sell this data to drug companies.

The resulting prescriber profiles—which include a physician’s name and the drugs he or she prescribes—have patient names removed by the time they are sold to drug companies. But armed with these profiles, marketers can tailor marketing material and detailing calls to individual providers. So, for example, if a doctor is prescribing Ambien for most of his sleep-deprived patients, he might receive marketing material and visits from Sepracor representatives that extol the benefits of Lunesta instead.

Critics see data mining as an invasion of privacy, for both patients and providers. From the provider side, many doctors are unaware that this confidential information is being sold. Ironically, the American Medical Association is a willing participant: the group sells physician profile information to companies like IMS to use in data mining efforts. In one year alone, the AMA profited $44 million in income from the practice (16% of its annual budget).

From the patient side, there is the concern that identifying information isn’t always removed by data mining firms before it is provided to marketers. Evidence of this comes from New Hampshire, where in 2006 the state legislature passed a prescription confidentiality law prohibiting the commercial use of prescribing records where the prescriber or patient is identified. According to New Hampshire Representative Cindy Rosenwald (one of the bill’s sponsors);

“We did hear testimony in New Hampshire that not all marketers actually do remove the patient’s name before they release those records to the outside vendor to begin the process of data mining the prescriber’s identity and prescribing record. . . .Our pharmacy board used to hear complaints from patients (about receiving ads in the mail for drugs to treat the medical conditions they have), and after our prescription confidentiality law went into effect, the complaints stopped.”

Ultimately, the kind of targeted marketing that results from data mining promotes the use of newer, more expensive drugs over their cheaper alternatives. It can lead to medication changes that harm patients and disrupt treatment. But from the industry’s point of view, data mining works. “According to the data mining industry itself,” states a Prescription Project fact sheet, “‘research has shown that winning just one more prescription per week from each prescriber yields an annual gain of $52 million in sales.’” 

The drug industry has stepped up these efforts even more;  they are now enlisting the help of large pharmacy chains and prescription drug benefit managers in marketing directly to doctors. Earlier this month, Daniel Carlat, a psychiatrist in Massachusetts who publishes The Carlat Blog, wrote that he received a mailing marked “Confidential” from CVS/Caremark, the pharmacy benefit manager and drugstore chain. The letter said that CVS/Caremark managed the benefits for one or more of Dr. Carlat’s patients, and that they were “committed to providing health care professionals with information about drug therapy.” This commitment apparently included providing Dr. Carlat with a 6-page newsletter about the benefits of prescribing Cymbalta (made by Eli Lilly) for fibromyalgia. Cymbalta, which had $2.7 billion in sales last year, was approved for use in treating fibromyalgia in June 2008.

“Pure and simply, it is an advertisement for Cymbalta, Eli Lilly's antidepressant which was recently approved for the treatment of fibromyalgia,” Carlat writes.

 "But it doesn't look like an ad. It looks like a letter from a pharmacy that is deeply concerned that my fibromyalgia patients receive the best treatment.”

Carlat goes on to question:

“How much is Eli Lilly paying CVS Caremark to perpetrate this deception? Which executive at CVS became so overcome with greed that he or she approached Eli Lilly about this joint venture? Has CVS Caremark informed its patients that it is selling their pharmacy information to a drug company?”

John Russell, a reporter for the Indianapolis Star decided to answer these questions and more this week in his investigative piece, “Lilly pays CVS Caremark to try to get doctors to prescribe Cymbalta.”
What he found out is that CVS Caremark is selling its services to other drug companies too. How much money the company is making on the venture is not public, but physicians around the country are reporting that they too are receiving mailings from CVS “that appear to look like important patient information but turn out to be promotions for Lilly, Merck, AstraZeneca, Bayer and other pharmaceutical companies,” according to Russell.

The letters can even refer to particular patients.

“Dr. Matthew Mintz, an internist in Washington, D.C., got a letter last fall from CVS Caremark, telling him he should consider prescribing the diabetes drug Januvia, made by Merck, for one of his patients taking a different medicine. The mailing was sponsored by Merck.

‘It's jus
t plain wrong,’ Mintz said. ‘It's an ad disguised as a professional-to-professional communication, using my patients' private data.’”

There is clearly a lot of creative thought that goes into designing new ways to market pharmaceuticals. And sometimes marketers step over the line. The industry has put out billions of dollars in settlements and fines for aggressively promoting drugs for off-label uses (Eli Lilly’s $1.4 billion settlement for improperly marketing Xyprexa, its antipsychotic, for unapproved uses, is the latest example) and for running deceptive ads that oversell benefits.

But as long as transparency rules remain mostly voluntary, companies will continue to employ new strategies, like data mining, to help garner more customers. Physicians will not be the only targets of these efforts. Websites designed to help connect patients with others suffering from the same disease—Patients Like Me is the leader in this area–sell data they’ve collected about medications taken by their members, for example, to drug companies. Companies that manage electronic health records for large health care organizations also sell “de-identified” patient data on diagnosis, medication, treatment, etc. to drug companies. The buying and selling of patient data, accessed through electronic health records or from disease websites or from pharmacies—it’s the fasted-growing method for better drug marketing.

It is time to crack down on prescription data-mining. The New Hampshire Prescription Privacy Law is the strongest statute limiting the practice, and so far it has been upheld in the face of legal challenges by IMS and Verispan on grounds that the law restricts free speech. In June the Supreme Court refused to review a lower court's decision to uphold the statute.

Vermont has passed a similar law that requires physicians to “opt-in” if they want their identities and information about their prescribing habits available for commercial use. Other states are considering similar legislation, and there are sure to be more court battles on this issue, especially as more industries employ marketing practices that depend on the buying and selling of consumer data.

In the end, there is clearly a greater need for transparency in the marketing practices of drug companies. If physicians and patients knew that their personal data was being manipulated to help companies sell drugs they, like Daniel Carlat, just might speak out about it. We also need to make a strong commitment to counter-detailing—putting together independent groups of experts who gather unbiased research on best treatment practices and visit doctors in their offices just like drug company reps do. The marketing geniuses will continue to find ways to sell more drugs—the challenge is to make sure these methods are ethically and legally sound—and to be astute enough to know when we’re being oversold.

16 thoughts on “Mining for Gold in Prescribing Records

  1. Like most companies, the pharmaceutical industry is not pure. While they have a nobile mission, they operative for profit, a word that has become a pejorative term to many. If these guys don’t make money, and a lot of it, then how will they conduct the R & D needed to develop tomorrow’s drugs. Most drugs never make it to market and those that do can get taken down after approval at any time, because of actual or suspected safety issues. If we want drug companies to take risks on real breakthroughs, such as a cancer vaccine, then they need an incentive to do so. Otherwise, they will crank out safe ‘me-too’ drugs that we don’t really need. These are not easy issues. As a physician, I am very distressed about data mining and other aggressive sales techniques. I take no $$$ from any of them, and I hope the free lunches I get haven’t corrupted me to the core. If we hit these companies hard,and they can’t invest in our health future, then does the public win? http://www.MDWhistleblower.blogspot.com

  2. Naomi, I’m sending your blog around the patient safety network and I’m calling for a boycott on CVS/Caremark. I cut up my CVS card and posted pictures on Facebook. That CVS/Caremark is using confidential medical records for purposes other than their intended is shameful, sickening, and downright creepy. They are taking money from the drug companies to harvest data from our sensitive, personal history… they are scumbags. Will patients EVER have any rights?

  3. Seems like I read somewhere last year that the much talked about R&D expenses no longer yield the great results they once did. The block-busters of the past were “low hanging fruit” from a research standpoint and the next big breakthroughs are too costly for even a well-heeled big company.
    Meantime, the tax-supported NIH is doing the heavy lifting. But thanks to some legerdemain from a few years ago those discoveries, once properly left in the public domain (tax money, you know) are now being peddled at wholesale rates to drug companies to milk maximum profits until the patents run out.
    Wish I could furnish links. Maybe someone else knows. Or maybe I’m misinformed.
    Anyone?

  4. Michael – The dollars would come from the sale of drugs, and the result would be that the public would pay much less for drugs and get more benefit from their purchases. Right now, drug purchases pay for profits, marketing expenses, advertising, market research, other overhead, “me too” drug development, plus a very modest leftover quantity for innovative new drugs. Under the FDDP plan I suggest, only the last item would be important – hence the substantial cost savings.

  5. The bipartisan Coalition for Patient Privacy got historic new patient rights into the stimulus bill—key is a BAN on the sale of each patient’s protected health information, which includes everything from DNA to prescriptions. See letter to Congress from the Coalition at: http://www.patientprivacyrights.org/site/DocServer/CoalitionPatPriv_Final01.14.09.pdf?docID=4701
    Act to save health privacy NOW and sign up for e-alerts at: http://www.patientprivacyrights.org

  6. Hello
    Its really interesting to read your post.Sometimes I feel that some of pharmaceutical companies are not genuine which is really not good.Thank you very much for giving such a good information.

  7. There are so many inaccuracies in the above post its hard to know where to start.
    For starters, none of this is new: prescription data has been around for two decades. Its primary use is for the drug companies to be able to segment physicians so they call on the right ones – much like a publisher would want a mailing list rather than blindly send out an offer. Despite the innuendo in this blog, there is no personally identifiable data collected on patients, nor is there any incentive to do so – and plenty of downside risk if they did. The example cited in the blog is not from a third party data company: its about CVS Caremark. a pharmacy that is entitled to have patient and physician information – though they are supposed to use it responsibly as part of providing patient care.
    Why this is suddenly news after 20 years is that physicians are increasingly afraid of being monitored by government- particularly in the growing health care push from Washington. All the ‘best practices’ language from D.C. is code for ‘we can see what you do and will start to challenge your decisions’. And they should – why shouldn’t doctors be accountable for doing the state-of-the art practices any more than your financial adviser or car mechanic for that matter. Yet, looking at the data far too many doctors don’t.
    Responsible data mining is a necessary component to better quality, lower cost health care we desperately need. If you’re not sure, just ask yourself this question: a family member has been diagnosed with a serious illness – would you like to know what treatment protocol has the best outcome and which physicians and hospitals have the best track record in treating it?

  8. Robert,
    Despite finding so many inaccuracies in my post that you hardly know where to start, you completely confuse drug company marketing with comparative effectiveness research. In fact, the kind of data mining I wrote about has absolutely nothing to do with “best practices” and everything to do with selling new, expensive drugs that may not be better than what a doctor is prescribing. If drug companies were really interested in “best practices” they wouldn’t pay pharmacy managers to try to disguise marketing material as educational literature. They wouldn’t “ghost-write” research articles that tout the benefits of their products for off-label uses or pay doctors to shill for them.
    Real “best practice” research involves studies that determine which drugs and treatment protocols are most effective–in resolving symptoms and in long-term consequences. There is a clear need for more of these studies, but they are almost never funded by drug companies who have little interest in pitting their expensive, proprietary new drugs against older generics or other treatments.
    It seems to me that most doctors resent having their prescribing information widely disseminated to marketers because, like anyone who receives piles of junk mail and phone calls from marketers, they don’t like the hard sell. Most would welcome real educational information.
    As far as confidential patient information being used by marketers, there have been many reports from New Hampshire and elsewhere of patients receiving drug company mailings about drugs targeted to their particular diagnosis.When laws are passed restricting data mining, these mailings ceased. That seems to suggest that confidentiality is not always maintained.
    Finally, the idea that “Responsible data mining is a necessary component to better quality, lower cost health care we desperately need” is really laughable. If by “responsible data mining” you mean the federal government, under the auspices of an expert committee responsible for conducting comparative effectiveness studies on say, the treatment of hypertension in the elderly, reviews physician prescribing records–then I agree. But since you really seem to be supporting unrestricted data mining by drug companies and their marketers in the interest of selling more drugs, I strongly disagree with this statement.
    Naomi Freundlich

  9. Robert:
    “would you like to know what treatment protocol has the best outcome and which physicians and hospitals have the best track record in treating it?”
    Yes, we would like to know this…where can we get this information? From drug companies who profit from selling drugs to us (or to our physician to prescribe to us)? That’s not a reliable source.

  10. Recently U.S news published “most influential doctors”, done by Qforma. I thought the list was so off ( atleast in my area). Most were private docs who saw volume of patients and probably had certian prescibing habits. They don’t reveal theeir methodolgy which sounds fishy. Why would u.s news publish this results?

  11. The Federal Government, or any other
    level of government is not a good example of “responsible data mining”. We may have dodged a privacy mortar shell with the
    probable failure of both the House & Senate bills. Trusting anything to government is incredibly naive, considering the historical record.

Comments are closed.