Voluntary Self-regulatory Codes: What Should We Expect?

Dana Katz, Arthur L. Caplan, and Jon F. Merz (2003) briefly describe the updated guidelines that the Pharmaceutical Research and Manufacturers of America (PhRMA) issued in May 2002 but fail to consider why those guidelines ended up looking the way that they did.

On the surface, voluntary self-regulatory codes from the pharmaceutical industry might look like a sensible approach to controlling the promotional activities of companies. Lacking government-industry resistance, such codes have the potential to be a more flexible and cost-effective option. In a highly competitive industry the desire of individual companies to prevent competitors from gaining an edge could be harnessed to serve the public interest through a regime of voluntary self-regulation run by a trade association (Ayres and Braithwaite 1992).

However, like many theories this one proves to be unsupported by the evidence. Lexchin and Kawachi (1996) point out that the mission of trade associations such as PhRMA is primarily to increase sales and profit. From the business perspective, self-regulation is mostly concerned with the control of anticompetitive practices. Therefore, when industrial associations draw up their codes of practice, they deliberately make them vague or do not cover certain features of promotion to allow companies a wide latitude. Many misleading advertising tactics are good for business. As a result, voluntary codes tend to be reactive, they lack transparency, they omit large areas of concern, and they lack effective sanctions.

Industry codes in Australia (Medicines Australia 2003), Canada (Rx&D 2003), and the United Kingdom (Prescription Medicines Code of Practice Authority 2001) all operate under a reactive as opposed to a proactive style of regulation. That is, action is generally taken only upon receipt of complaints; it does not attempt to prevent breaches from occurring in the first place. The Australian and British industry associations claim to monitor compliance with their voluntary codes by scrutinizing a random selection of advertisements in medical journals. However, their monitoring procedures are not defined beyond vague statements like this one from the Australian code: it states that it will "monitor selected promotional material ... on a regular and ongoing basis" (Medicines Australia 2003, 114). Because there is no adequate monitoring of the activities of sales representatives in Australia, these individuals provide only minimal information on product risk. Thirteen of 16 presentations contained at least one inaccuracy when compared with Australian Approved Product Information, and presentations did not always comply with the then current guidelines in the Medicines Australia Code of Conduct (Roughead, Gilbert, and Harvey 1998).

A second feature common to many voluntary codes is their lack of transparency. For example, although Medicines Australia "will on occasion, provide information regarding the activities of the Code of Conduct Committee to the general public including media outlets" (Medicines Australia 2003, 122), it will not release any information before the relevant company has been contacted. The code does not say what will happen if the company refuses to allow information to be released. The Canadian code contains no provisions for a review of its operations or for consulting with individuals or organizations outside the industry when the code is being updated.

A recent review of journal advertisements in Australia found that in 174 distinct advertisements, only 13 reported quantitative statistics about outcomes and none used absolute risk reductions (ARRs) or number needed to treat (NNT), while ten used relative risk reductions (RRRs) (Loke, Koh, and Ward 2002). Reporting outcomes as RRRs as opposed to ARRs or NNTs increases physicians' willingness to prescribe a product (Naylor, Chen, and Strauss 1992). The Medicines Australia code contains no provisions mandating the use of ARRs or NNTs. While the Rx&D code devotes three of its 15 pages to rules about continuing-health education, the code does not oblige companies to disclose sponsorship of programs or commercial relationships with speakers. There is no requirement that generic names be used in presentations, and there is nothing to prevent the distribution of promotional items at...