Advertisement
   << Issue   << Category        

Data Mining Revisited—Again and Again

Jesse C. Vivian, RPh, JD
Professor, Department of Pharmacy Practice
College of Pharmacy and Health Sciences
Wayne State University

Detroit, Michigan



1/20/2011

US Pharm
. 2011;36(1):68-70. 

Over the past several years, there has been much ado in the pharmacy world over whether the conduct of data mining is lawful or not. Data mining is the process whereby companies that want to collect data about practioners’ prescribing habits contract with pharmacy organizations (primarily the larger chain-store companies) to search their patient computer records to find what is called prescriber-identifiable (PI) data.1 The procedure is supposed to scrub patient-identifiable information from the collected data so as to not commit any Health Insurance Portability and Accountability Act (HIPAA) violations. The question is whether the collected information is an invasion of privacy of the prescriber or simply information that should be made publicly available, at least to those companies wanting to pay for it, as a tool for pharmaceutical companies to tailor their marketing efforts to certain prescribers in order to encourage more prescribing of certain drugs. 

The Data Mining Business

Just to put this practice into perspective, the two leading purchasers of this data, IMS Health and Verispan, together monitor over one billion prescriptions per year. Approximately 51,000 pharmacies in the United States are wired to facilitate data mining, resulting in revenues to the “data miners” in excess of $2 billion annually. In addition, the pharmaceutical industry spends more than $30 billion annually on detailing, drug samples, ads in medical journals, and direct-to-consumer advertising, with approximately $7 billion spent on direct marketing to physicians (averaging $9,000 per prescriber). The American Medical Association (AMA) also sells data to the groups that gather this information and, in one year alone, earned $44 million from this practice.2 

Although the data mining bans passed by two states (New Hampshire and Maine) were determined by the First Circuit Court of Appeals to be constitutional, meaning that the data mining companies were put out of business—at least as to this process—in those two states,3 the Second Circuit Court of Appeals recently ruled that Vermont’s ban on data mining is a constitutionally impermissible restriction on commercial speech.4 The Vermont decision is diametrically opposed to the New Hampshire and Maine cases. In other words, the First Circuit and Second Circuit Courts of Appeals have decided essentially the same question with directly opposing opinions. When something like this happens, the potential for final resolution is ripe for review by the U.S. Supreme Court, if any of the parties push the “court of last resort” into hearing arguments. When there are splits like this one, between different circuits in the various Courts of Appeals, the Supreme Court is most likely to resolve the constitutional dispute if it is persuaded that the issues are important enough for it to consider. Put another way, don’t be surprised if the Supreme Court does decide to make a final ruling in these cases. 

The Vermont Case

The state of Vermont adopted a statute that bans “the sale, transmission, or use of ‘PI data’ for marketing or promoting a prescription drug unless the prescriber consents.”5 Following the passage of the statute (hereafter referred to as §17), the plaintiffs, which included IMS Health, Verispan, Source Healthcare Analytics, and the Pharmaceutical Research and Manufacturers of America (PhRMA), sued in federal court seeking declaratory and injunctive relief and moving for summary judgment, challenging the act as an impermissible restriction on commercial speech and in conflict with the so-called “dormant” Commerce Clause of the U.S. Constitution. The trial court denied the motions, finding that “the Vermont statute is a constitutionally permissible commercial speech restriction.” The Second Circuit reversed the trial court’s findings, essentially repealing Vermont’s ban on data mining.6 

On appeal, the state argued that even if §17 restricts commercial speech, it is narrowly tailored to directly advance the state’s interest in protecting medical privacy, controlling health care costs, and promoting public health. However, the court held that §17 of the Act is “a commercial speech restriction that does not directly advance the substantial state interests asserted by Vermont, and is not narrowly tailored to serve those interests, [thus] the statute cannot survive” the prevailing standards for analysis of this kind of case. The prevailing standards are often referred to as the Central Hudson test, named for a case in which the Supreme Court set forth the standards for determining whether “commercial speech” is permitted under the Constitution.7 

In its analysis of the issues, the court cited plaintiffs’ claims that the data-mining process analyzes prescription information sold to them by pharmacies, stripped of PI information, and that they sell this analyzed data, revealing individual physician prescribing patterns, to PhRMA companies and other interested parties. PhRMA companies, they claim, “use PI data to identify audiences for their marketing efforts, to focus marketing messages for individual prescribers, to direct scientific and safety messages to physicians most in need of that information, to track disease progression, to aid law enforcement, to implement risk mitigation programs, and to conduct clinical trials and post-marketing surveillance required by the…FDA.” The court commented that “while §17 in part aims to decrease detailing, prescribers may want to receive the information detailers provide, and, in any event, prescribers are free to decline meetings with detailers.” 

The court noted that manufacturers are not the only entities that purchase PI data from the data-mining appellants, but they are the only customers banned from using PI data in their marketing efforts by §17. Other users of PI data include the state, researchers, the FDA, the CDC, the Drug Enforcement Agency (DEA), and insurance companies. The data-miner plaintiffs contend that sales to manufacturers covered by §17 are essential to enable them to provide PI data for these other, permitted uses. 

The court reviewed the litigation history of the comparable New Hampshire and Maine statutes, noting in part that neither survived intermediate scrutiny at the district court level, but that the First Circuit reversed the District of New Hampshire Court8 and followed this holding by also reversing the District of Maine Court decision.9 In reviewing the Vermont statute, the court found that generally unlike Maine and New Hampshire, the Vermont Legislature issued 31 findings in support of the statute, expressly stating the legislature’s intent to interfere with the marketplace of ideas to promote the interests of the state. Examples included the legislature’s view that the goals of marketing drug products are “often in conflict with the goals of the state” and that the “marketplace for ideas on medicine safety and effectiveness is frequently one-sided,” such that prescriptions are issued based on incomplete and biased information. The legislature, the court observed, found that public health is ill-served by the massive imbalance in information presented to doctors and other prescribers. The court concluded that §17 is the state’s attempt to correct what it sees as an unbalanced marketplace of ideas that undermines the state’s interests in promoting public health, protecting prescriber privacy, and reducing health care costs. 

The Vermont attorneys argued that the statute is not a restriction on speech, but on a commercial practice. They claimed that data miners buy and sell a commodity and this activity can be regulated. The state conceded that the PhRMA members who seek to use that commodity present a “closer question under the First Amendment,” but “the statute is nevertheless a restriction on commercial conduct.” The court agreed with the district court for the notion that the First Amendment protects a wide variety of speech and for the proposition that “speech in a form that is sold for profit is entitled to First Amendment protection.”10 But it held that the Vermont law, and the legislature’s intent, is not merely a restriction on commercial practice; it also affects commercial speech. 

The Central Hudson Test

As such, the court ultimately determined it should apply the four-step process outlined as the Central Hudson test. As to the first prong, it found that there was no contention that the speech involved was misleading or related to unlawful activity. Thus, only the remaining three prongs of the Central Hudson test were considered. 

On the second prong of the Central Hudson issues in dispute, the court held that there was no substantial controversy whether the state’s interests in promoting public health and cost containment were substantial state interests. However, the parties disagreed as to whether protecting the privacy of prescribers is a substantial state interest, particularly given the court’s recognition that “the statute plainly does not protect physician privacy,” since multiple uses of PI data are allowed that are not related to the marketing of drug products. Rather, the court held, the medical privacy concerns are comprised of two distinct interests. The first is the integrity of the prescribing process itself, and the second is preserving patients’ trust in their doctors by preventing patients from believing that their physicians are inappropriately influenced by PI data-driven marketing. The court found these interests “too speculative to qualify as a substantial state interest.” 

Turning to the third Central Hudson prong, the court found that the statute “cannot be said to advance the state’s interests in public health and reducing costs in a direct and material way.” Furthermore, the statute seeks to regulate conduct the state opposes by restricting speech. The legislative findings are explicit that Vermont here aims to do exactly what the Supreme Court in Central Hudson and its prodigy has so highly disfavored, i.e., to put the state’s thumb on the scales of the marketplace of ideas in order to influence conduct. “In other words,” stated the court, “the statute seeks to alter the marketplace of ideas by taking out some truthful information the state thinks could be used too effectively.” 

The court also held that the statute failed the fourth Central Hudson prong, if for no other reason than that it restricts the use of PI data in marketing all brand name drugs, whether or not there is a generic alternative. The statute bans speech beyond what the state’s evidence purportedly addresses. 

One member of the panel dissented. Her summary opening statement opined that by “misconstruing Vermont’s prescription confidentiality law…as a direct restriction on pharmaceutical marketing…the majority extends First Amendment protection to data miners and pharmaceutical companies principally challenging a restriction on access to otherwise private information. In so doing, the majority not only reaches the wrong result in this case, but creates Circuit precedent likely to have pernicious broader effects in a complex and evolving area of First Amendment law.” 

Analysis

There is no question that this is a difficult case to understand, given the reasoning and language of the majority opinion. Even more confusing is the logic used by this court to distinguish the findings of the First Circuit’s holdings in the Maine and New Hampshire cases, holding that states have the right to limit commercial speech even if that speech also constitutes commercial conduct. As demonstrated in these cases, the distinction between commercial speech and conduct depends on a very narrow line that does not necessarily carry with it a neat distinction on how one views these matters. 

It almost seems as if the Second Circuit in this case was determined to disagree with the reasoning of the judges in the First Circuit just to create a controversy that must be resolved by the Supreme Court. That would be a very disingenuous reason to hold Vermont’s statute unconstitutional, although stranger things have happened in the world of jurisprudence. But for now, the only means of resolving these different points of view rests with the Supreme Court if and when it agrees to hear appeals from the involved parties. It’s a pretty safe bet that the Supreme Court will take on the cases to provide a binding answer to the questions presented. For now, it is just a waiting game to see when that will happen. 

REFERENCES

1. Vivian JC. Pharmacists beware: data mining unconstitutional. US Pharm. 2009;34(6):48-49. www.uspharmacist.com/content/ d/pharmacy_law/c/13856/. Accessed December 20, 2010.
2. See Note 1, supra.
3. See Note 1, supra.
4. IMS Health v. Sorrrell, Slip Op No. 09-1913-cv(L), 09-2056 (CON) November 23, 2010. http://caselaw.findlaw.com/us-
2nd-circuit/1545632.html. Accessed December 20, 2010.
5. Vt Acts No. 80 §17 (2007), Vt Stat Ann 18 §4631.
6. See Note 4, supra.
7. Central Hudson Gas & Electric Corp. v. Public Service Commission, 447 U.S. 557 (1980).
8. IMS Health Inc. v. Ayotte, 550 F3d 42, 2008.
9. IMS Health Inc. v. Mills, 616 F3d 7 (1st Cir), 2010.
10. Citing, inter alia, Va State Bd of Pharmacy v. Va Citizens Consumer Council, 425 U.S. 748, 1976. 

To comment on this article, contact rdavidson@uspharmacist.com.

Popular Articles
Advertisement