Earlier this summer the Royal College of Paediatrics and Child Health published a report describing its policies towards accepting industry sponsorship.1 The ethics advisory committee that wrote the report was formed in response to controversy surrounding the college’s acceptance and lack of disclosure of sponsorship from Nestlé, a manufacturer of breast milk substitutes. The acceptance of the money from Nestlé was hotly debated among college members because breast milk substitutes are associated with infant deaths in developing countries and do not provide the same health advantages as breast feeding. The college voted 73% in favour of continuing to accept sponsorship from baby food manufacturers but recommended that it should define criteria for ethical sponsorship. An examination of the college’s resulting report and its recommendations may offer some help to other organisations struggling with the same issue.
Overwhelming evidence exists that single source sponsorship is associated with outcomes favourable to the sponsor’s product.2–4 Although most documentation of industry influence on research concerns the pharmaceutical and tobacco industries, other types of corporate sponsors are also known to influence research reports.5,6 One reason why published research favours the sponsor’s product is because sponsors sometimes suppress publication of unfavourable findings.7 Single source sponsorship can also influence decisions, such as prescribing decisions, that are more directly related to patient care.8 In addition to the empirical data on the influences of industry sponsorship, doctors’ acceptance of money and gifts from corporate sponsors creates a relationship with the sponsor. Doctors may then feel favourable towards, or even obligated to, that sponsor.9,10
The Royal College of Paediatrics and Child Health’s report states that the college’s guiding principle is that all its activities should be in the best interests of children worldwide. However, it is also concerned about the public perception of accepting commercial sponsorship—and the desire to produce favourable public perceptions may indeed be the driving force behind these recommendations. The recommendations for reducing both the real and the perceived influences of industry funding include designating sources of sponsorship as acceptable or not and, if acceptable, establishing restrictions on sponsorship.
The report attempts to differentiate sponsors that are acceptable from those that are not, yet these distinctions are slippery. Sponsorship from or investment in any company which produces tobacco, manufactures arms, or exploits children is unacceptable, while sponsorship from companies which market pharmaceutical products, medical equipment, or mineral water is deemed acceptable. The college has based this distinction on how the products are used or marketed and not on the products themselves. However, assessing the ethical practices of companies in an international market with frequent mergers and acquisitions is a formidable task. Companies falling into a middle category include those manufacturing alcohol, sweetened drinks, and breast milk substitutes, and sponsorship by such companies is likely to remain controversial. The college might therefore be better off focusing on its second strategy of establishing restrictions on sponsorship, regardless of its source.
The college recommends a series of restrictions that apply to any acceptable sponsorship. Firstly, all sponsorship must be fully and transparently recorded in the college accounts and should be disclosed on the college website, by the units and departments accepting money, and in educational materials. Full disclosure of industry support forces individual members of the college to ask whether “they would be happy for it to be generally known that they are receiving sponsorship.”1 However, disclosure is not a panacea. Disclosure does not necessarily eliminate the influence of industry funding on research or doctors’ behaviour. Moreover, patients and the public may distrust organisations or doctors who are funded by unpopular sources.11 Lastly, disclosure may be difficult to enforce, as suggested in a recent study showing that 70% of articles from journals with disclosure policies made no mention of potential conflicts.12 The report’s recommendation that a committee should be established to monitor the information given about sponsorship by the members of the college is, however, a step towards ensuring that accurate disclosure occurs.
Secondly, sponsorship from companies producing breast milk substitutes or others in the middle category is not acceptable for general college activities but is acceptable for named activities such as travelling fellowships. Therefore, individuals can choose not to accept funds derived from baby food manufacturers. This “ethics shifting” (A Shulz, personal communication) is comparable to cost-shifting in economics—where the burden (in this case, the unethical decision) is transferred from the group to the individual. In effect the organisation avoids making a decision.
Thirdly, the college recommends that sponsorship for individuals should be modest, although a lower limit for sponsorship that lacks influence has not been established.13 The cumulative amount of sponsorship, as well as each individual amount, should be taken into account.
Fourthly, the college recommends that sponsored investigators should retain control over the publication of results, regardless of their outcome. This important restriction could help to eliminate some of the influence of single sponsors over research outcomes.
If the college enforces its restrictions on sponsorship, and considers extending them to all corporate sponsors, it will set a strong example for other organisations that must deal with the reality of accepting increasing corporate sponsorship while maintaining ethical standards.