Mining Ban Unconstitutional

US Pharm. 2007;32(6):67-70.

The New Hampshire Legislature enacted a law that bans pharmacies, insurance companies, and similar entities from transferring or using prescriber-identifiable data for certain commercial purposes.¹ Violations of the statute are punishable as a misdemeanor if the offender is a person and are treated as a felony if the wrongdoer is any other legal entity such as a corporation; violators are also subject to civil penalties. ² The legislative history indicates that: "The legislature passed the law to protect patient and physician privacy and to save the State, consumers, and businesses money by reducing health care costs by encouraging the use of generic drugs as opposed to brand name and newer, more expensive medications. New Hampshire is the only state that has enacted such legislation to date." Several other states are considering introducing similar measures.

Lawsuit
Two organizations that use data of this type filed a lawsuit contending that the law impermissibly restricts their First Amendment right to free speech. ³ They asked a federal district court to enter an injunction prohibiting the state from enforcing the law. The plaintiffs IMS and Verispan are considered to be the world's leading providers of information, research, and analysis to the pharmaceutical and health care industries. IMS, which is considered the largest operator in the field, purchases prescriber information from approximately 100 different suppliers. Verispan, the other plaintiff, is a company thought to be only about one tenth the size of IMS, obtains its information from about 30 to 40 suppliers. The majority of these suppliers are pharmacies that receive payments from the data miners. The record showed that these two organizations collectively acquire and analyze data from billions of prescription transactions per year in the United States.

In reviewing this matter, the federal district court judge authoring the opinion noted:

A lucrative market has developed in recent years for data identifying the prescribing practices of individual health care providers ("prescriber-identifiable data"). Pharmacies acquire prescription data in the ordinary course of business. Data mining companies … purchase the prescription data, remove information identifying patients before it leaves the pharmacy, combine what remains with data from other sources, and sell the combined data to interested purchasers. The data miners' biggest clients by far are pharmaceutical companies, which use the data to develop marketing plans targeted to specific prescribers.⁴

One way that pharmaceutical manufacturers use this data is to identify "early prescribers" of newly marketed medications. This, of course, is a way for companies that have expended millions of dollars in research and clinical marketing prior to obtaining a premarket clearance from the FDA to get a legitimate quick start on recovering research and development dollars. Another use of this data is for the drug makers to develop marketing strategies and detailing ideas for pharmaceutical representatives to "detail" targeted physicians.

The judge's written opinion containing his ruling includes some fascinating statistics. There are approximately 1.4 million licensed health care providers authorized by state laws to issue prescriptions in the U.S. for approximately 8,000 different prescription-only products in various forms, strengths, and doses. These prescriptions are filled by approximately 54,000 retail pharmacies and other licensed medical facilities throughout the country. Adding a little more statistical data not relevant to the case, but certainly interesting, these numbers suggest that each physician supports about 26 pharmacies.⁵

According to the judge, the system works like this:

Retail pharmacies acquire prescription data during the regular course of business. For each prescription filled, a record is kept that includes the name of the patient, information identifying the prescriber, the name, dosage, and quantity of the prescribed drug, and the date the prescription was filled. If the pharmacy is part of a larger organization with multiple retail outlets, each outlet's prescription data is ultimately aggregated with data from other outlets and stored in a central location.⁶

In compliance with state and federal laws regarding patient privacy, the pharmacies that participate in the data-mining process allow the companies doing the information gathering to install software on their computers that encrypts any information identifying patients before it is transferred to their computers. 

This "prescriber-identifiable" data is combined with prescriber reference information that permits the data miners to match prescriptions to the correct prescriber, supply the prescriber's correct name, and add address, specialty, and other professional information about the prescriber to the prescription data. These prescriber reference files are obtained from sources that include the American Medical Association's (AMA) Physician Masterfile. The AMA's Masterfile contains demographic, educational, certification, licensure, and specialty information for more than 800,000 active U.S. medical doctors and over 90% of osteopathic doctors.

This information, as might be imagined, is extremely important to the marketing strategies of the pharmaceutical and related industries. According to evidence, substantially all of the revenue generated by the two companies in 2005, 2004, and 2003 involved sales of prescriber-identifiable data to pharmaceutical producers. The information is also sold to biotechnology firms, pharmaceutical distributors, government agencies, insurance companies, health care groups, researchers, consulting organizations, the financial community, manufacturers of generic drugs, pharmacy benefit managers, and others. Some of the purchasers of this data use, license, sell, or transfer the information for advertising, marketing, and promotional purposes.⁷

There should be little doubt that these data are crucial to physician-aimed marketing tactics. For example, the judge noted that in 2000:

[T]he pharmaceutical industry spent approximately $15.7 billion on marketing, $4 billion of which was dedicated to direct-to-physician strategies. More recent estimates suggest the industry currently spends between $25 billion and $30 billion per year on marketing. The large pharmaceutical companies spend roughly 30 percent of their revenues on promotion, marketing, and administration, while spending only approximately 13 percent on research and development."

The judge noted that drug makers depend on a variety of direct marketing techniques to reach health care providers, including efforts to enlist the support of "thought leaders," characterized as "physicians and researchers whose views are accorded special weight in the medical community." He went on to observe that pharmaceutical companies enlist the support of these individuals by sponsoring their research, retaining them to serve as consultants and speakers, and entertaining them at dinners and other events. He also stated that the tacit support of thought leaders is deemed by pharmaceutical companies to be highly valuable in persuading others to prescribe their products.

Insofar as face-to-face or one-on-one "detailing" goes, the judge noted that sales representatives often provide prescribers with both written and oral information about particular drugs in an effort to persuade them to prescribe the drugs being detailed and that they also offer prescribers free samples that can then be distributed to patients at no charge. He wrote that certain reports estimate that the total annual retail value of sampled drugs exceeds $11 billion. He went on to claim that because some prescribers are reluctant to meet with sales representatives, "small gifts, free meals, and other inducements are also frequently offered to health care providers and their staffs in an effort to facilitate access and encourage receptivity to the representative's sales pitch." The judge asserted that in addition to facilitating access, "such inducements help sales representatives build relationships with prescribers that can make them more receptive to the product information that sales representatives provide."

Outcome
For a variety of reasons, the judge granted the plaintiffs' request for an injunction by holding that the New Hampshire law banning the use of physician prescribing data is an unconstitutional violation of the First Amendment's guarantee of free speech. He concluded that, "The challenged law restricts the transmission of truthful information concerning the prescribing practices of New Hampshire's health care providers [and that] it is not exempt from First Amendment review merely because it targets factual information rather than viewpoints, beliefs, emotions, or other types of expression." He also opined that the statute in question "restricts speech by preventing pharmaceutical companies from using prescriber-identifiable information both to identify a specific audience for their marketing efforts and to refine their marketing messages." In coming to this conclusion, the judge reasoned that the statute's primary purpose "is to affect commercial transactions by making it more difficult for pharmaceutical companies to convince health care providers to prescribe brand-name drugs when less expensive and equally effective alternatives are available."

New Hampshire argued that it has a substantial interest in protecting prescriber privacy by "limiting unwarranted intrusions into the decision-making process of prescribing physicians." The judge responded that the statute does not promote public health or influence health cost savings, even though he did accept the basic premise that pharmaceutical companies use prescriber-identifiable data to make detailing more persuasive. In his opinion, "Any general claim that the public health is undermined when the effectiveness of detailing for brand-name drugs is increased depends upon the counterintuitive and unproven proposition that, on balance, brand-name drugs are more injurious to the public health than generic alternatives." When the state claimed that it should be entitled to ban the use of prescriber-identifiable data because it is being used to target "early adopters" for the marketing of dangerous new drugs, the judge determined that the evidence "did not establish either that early adopters are more likely to be influenced by detailing than other health care providers or that new drugs are generally more injurious to the public health than existing medications." Accordingly, he concluded that the state did not show the law promotes public health.

The state also tried to advance the notion that the statute directly promotes its interest in containing health care costs. But this assertion, the judge said, depends on the assumption that any health care cost savings that will result from a ban on the use of prescriber-identifiable data can be achieved without compromising patient care and that less expensive generic drugs are just as effective as the more expensive brand name drugs. In this judge's mind, "This proposition is far from self-evident [because] non-bioequivalent generic drugs are not always as effective as brand-name alternatives." The judge did not stop here. He went on to state, "Even in cases where non-bioequivalent generic drugs will work as well or better than a brand-name alternative for most patients, there may be some patients who will benefit by taking the branded medication.
Yet, a ban on the use of prescriber-identifiable data affects both helpful and harmful brand-name prescribing practices in the same way."

In conclusion, he wrote:

Because the state failed to prove that any reductions in health care costs that may result from a ban on the use of prescriber-identifiable data can be achieved without compromising patient care, I am unable to endorse [the] argument that the Prescription Information Law can be justified as a cost containment measure.

After finding the law unconstitutional, the judge offered some pointers as to how the state might reduce health care costs without offending the constitution by restricting protected speech:

First, if legislators are concerned that pharmaceutical companies are improperly using samples, gifts, meals, and other inducements to promote inadvisable prescribing practices, they can address this perceived problem by following other states that have adopted laws that limit such practices.⁸

Second, if legislators fear that pharmaceutical detailing is simply too effective to go unrebutted, they can require the State to enter the intellectual marketplace in several different ways with competing information that will help health care providers balance and place in context the sales messages that detailers deliver. Among other things, they can require the State to prepare and distribute "best practice" guidelines that educate health care providers as to both the health and cost implications of their prescribing decisions; require the State to develop counter-detailing programs that make health care providers aware of the cost implications of their prescribing decisions;⁹ or they can require health care providers to regularly participate in continuing medical education programs that are specifically designed to provide practitioners with the best available information concerning the advantages and disadvantages of prescribing generic drugs rather than brand-name drugs.

Finally, if legislators are concerned that pharmaceutical companies are using prescriber-identifiable data to drive up Medicaid drug costs, they can address the issue directly by properly implementing a Medicaid Pharmacy Program that takes into account the cost-effectiveness of brand-name drugs when compared with nonbioequivalent generic alternatives. New Hampshire's Medicaid Pharmacy Benefit Program requires health care providers to obtain authorization from state officials before prescribing certain drugs for Medicaid patients.¹⁰ The State has also adopted regulations that both authorize the State to take cost considerations into account when deciding which drugs should be subjected to the prior authorization requirement,¹¹ and permit the State to reject requests to prescribe drugs that are subject to prior authorization.¹² Accordingly, the State can prevent unnecessary expenditures on brand-name drugs simply by subjecting such drugs to prior authorization and rejecting requests to prescribe them when they are not medically necessary.

Analysis
This is an opinion from a federal district court judge sitting in New Hampshire. The Attorney General of that state has indicated that an appeal will be made to the next higher court. Do not be surprised if it winds up on the doorstop of the U.S. Supreme Court. For the moment, all this decision means is that the status quo that existed prior to the state's attempted ban on data-mining procedures is preserved. The plaintiffs can continue with business as usual.

As far as its implications to pharmacy, the case also means that pharmacies that permit data mining of the information that gets stored in computer records may continue participating in these activities. But do not get complacent. As indicated, several other states are anticipating adopting legislation of this type. Further, the outcome of this case could be reversed at some future point. Obviously, this warrants some attention to developments in this arena.

The other piece of information that may be troubling for pharmacists is the judge's suggestion to expand the use of "prior authorization" procedures in Medicaid programs. This kind of approach to the cost-containment goals of states can shift a large burden of time and expenses to pharmacists if a physician prescribes a drug subject to these requirements without getting the state's permission.

References
1. See 2006 N.H. Laws § 328, codified at N.H. Rev. Stat. Ann. §§ 318:47-f, 318:47-g, 318-B:12(IV) (2006) (Prescription Information Law). The Prescription Information Law became effective on June 30, 2006 and is codified at N.H. Rev. Stat. Ann. §§ 318:47-f[*20] , 318:47-g, 318-B:12(IV) (2006). The statute states: "Records relative to prescription information containing patient-identifiable and prescriber-identifiable data shall not be licensed, transferred, used, or sold by any pharmacy  benefits manager, insurance company, electronic transmission intermediary, retail, mail order, or Internet  pharmacy  or other similar entity, for any commercial purpose, except for the limited purposes of  pharmacy  reimbursement; formulary compliance; care management; utilization review by a health care provider, the patient's insurance provider or the agent of either; health care research; or as otherwise provided by law. Commercial purpose includes, but is not limited to, advertising, marketing, promotion, or any activity that could be used to influence sales or market share of a pharmaceutical product, influence or evaluate the prescribing behavior of an individual health care professional, or evaluate the effectiveness of a professional pharmaceutical detailing sales force."
2. N.H. Rev. Stat. Ann. § 318:55.
3. IMS Health Inc, v. Ayotte (Attorney General of New Hampshire, Slip Op No 06-cv-280-PB, Opinion No. 2007 DNH 061 P (April 30, 2007), USDC, NH, 2007 U.S. Dist. LEXIS 31779.
4. Id.
5. 1.4 million prescribers divided by 54,000 pharmacies.
6. See Note 3.
7. It was noted that the "Plaintiffs also make prescriber-identifiable data available at little or no cost for non-marketing purposes to academic researchers, medical researchers, humanitarian organizations, and law enforcement authorities. These entities use the information to track patterns of disease and treatment, conduct research and clinical trials, implement best practices, and engage in economic analyses." See Id, footnote 2 of the opinion.
8. See, e.g., Minn. Stat. Ann. § 151.461 (2007) and Cal. Health and Safety Code § 119402(d)(1) (2007).
9. See, e.g., W. Va. Code Ann. § 5-16C-9(5) (2006) (authorizing state to develop counter-detailing programs).
10. See generally, 2004 N.H. Laws, ch. 188 (authorizing the New Hampshire Department of Health and Human Services to establish a preferred drug list and a prior authorization process).
11. N.H. Admin. Rules, He-W 570.06(f)(3),
12. N.H. Admin. Rules, He-W 570.06(I)-P).

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