Industry Support of Patient Advocacy Organizations: The Case for an Extension of the Sunshine Act Provisions of the Affordable Care Act

Abstract

Patient advocacy organizations (PAOs) have long been regarded as important representatives of patient and caregiver interests in health policy debates. Recently, however, PAOs have attracted increased scrutiny over their financial ties to drug and device companies. In the past year, researchers and policymakers have called for the creation of a “sunshine law” requiring mandatory public reporting of industry payments to PAOs. Others have suggested that increased transparency would do little to address, and may even exacerbate, underlying concerns about proindustry bias among industry-funded PAOs.

To date, however, the benefits of a sunshine law have not been well articulated, nor have objections to the idea been carefully addressed. In particular, little attention has been paid to clarifying the merits of statutorily mandated disclosure relative to those of increased voluntary disclosure by PAOs.

I examine arguments for and against a sunshine law and conclude that the balance of reasons supports the enactment of such a law.

In September 2016, at the height of the public furor over Mylan’s decision to hike the price of its epinephrine autoinjector EpiPen, The New York Times reported that the drugmaker was quietly orchestrating a campaign to have EpiPen added to the federal preventative drugs list.¹ If successful, the campaign would have eliminated copayments for EpiPen, appeasing Mylan’s most vocal critics while leaving government agencies, employers, and private insurers to shoulder the high cost of the product.¹

The public faces of the campaign included a physician who had been a paid consultant to Mylan and a coalition of patient advocacy organizations (PAOs), led by the Allergy & Asthma Network. Thanks to the Centers for Medicare & Medicaid Services’ (CMS’s) Open Payments Database (OPD), which the Times cited in its story, the scope and history of Mylan’s financial relationship with the physician was clear—down to the last cent. But the details of Mylan’s arrangement with the PAOs were far murkier. Although Mylan and the Allergy & Asthma Network acknowledged a financial relationship, they declined to disclose specifics.

This discrepancy in transparency reflects an underlying legal reality. The Sunshine Act provisions of the Affordable Care Act require drug and device manufacturers to report payments to physicians and teaching hospitals. These payments are cataloged in the publicly accessible OPD. However, there is no requirement that drug and device companies report payments to PAOs. Thus, even as PAOs work to influence policy in ways with far-reaching implications for the health care system, there is often little clarity about their financial backing.

One way to ensure greater accountability for PAOs would be to extend the federal sunshine law to cover industry payments to PAOs. A version of this idea was proposed in the Institute of Medicine’s 2009 report on conflicts of interest (COIs) and has been echoed in years since.^2–5 The idea came to the fore again recently with reports that Senator Claire McCaskill (D, MO) plans to introduce a bill that would make industry payments to PAOs subject to disclosure requirements.⁶ To date, however, the benefits of extending the sunshine law to PAOs have not been well articulated, nor have objections to the idea been carefully addressed.

BACKGROUND

If financial associations like the one between Mylan and the Allergy & Asthma Network were rare and transparent, a sunshine law would perhaps be unnecessary. But as recent studies have shown, industry payments to PAOs are extremely common and often opaque (Table 1). Even so, there would remain little need for a sunshine law if such relationships did not also raise ethical concerns. Some have denied that they do, suggesting that industry and PAO interests are aligned rather than in conflict.¹¹ But this argument is misguided. PAOs and drug and device companies may indeed find common cause on certain issues, but they have fundamentally different objectives. PAOs’ objective—at least in principle—is advancing the interests of the patients and caregivers. Drug and device companies’ objective is generating profits for owners and shareholders. Because there is no guarantee that what is good for patients and caregivers will be good for owners and shareholders, PAOs that accept industry funding are likely to face decisions that pit the interests of their funders against those of their constituents.

TABLE 1—

The Extent and Disclosure of PAO–Industry Ties: Findings From Recent Studies

Study	Sample	Methods	PAOs With Financial Ties to Drug, Device, or Biotechnology Companies, %	Findings Related to Disclosure of Financial Ties
Abola and Prasad7	PAOs for specific cancer subtypes recommended by the National Comprehensive Cancer Network (n = 68)	Review of PAO Web sites	≥ 75	76.0% of PAOs reported whether they had biopharmaceutical sponsorship
Claypool, Public Citizen8	PAOs that signed public letters opposing a Medicare Part B demonstration project (n = 147)	Review of PAO Web sites and annual reports	≥ 75	Study did not investigate the extent of PAOs’ disclosure
Hilzenrath, Project on Government Oversight9	PAOs that participated in the FDA’s PDUFA VI discussions (n = 42)	Review of PAO Web sites	≥ 93	Study did not investigate the extent of PAOs’ disclosure
McCoy et al.5	US-based PAOs with annual revenue ≥ $7.5 (n = 104)	Review of PAO Web sites and annual reports	≥ 83	88.0% of PAOs disclosed names of donors; 52.0% of PAOs disclosed donation amounts in ranges; 5.0% disclosed donation amounts in exact figures; 10.0% disclosed the uses of individual donations
Rose et al.10	Nationally representative random sample of US-based PAOs (n = 245a)	Survey of PAO leaders	67b	73 of 284 respondents (25.7%) reported that their organizations have policies for public disclosure of financial relationships
Rothman et al.4	PAOs awarded grants from Eli Lily during the first 2 quarters of 2007 (n = 188)	Review of PAO Web sites and annual reports	Sample included only PAOs that had received grants from Eli Lilly	25.0% of PAOs acknowledged Lilly’s contributions on their Web sites; 18.0% acknowledged Lilly’s contributions in their annual report; no PAOs disclosed the exact amount of a Lilly grant

Note. FDA = US Food and Drug Administration; PAO = patient advocacy organization; PDUFA = Prescription Drug User Fee Act.

^aNumber of respondents who provided information about their funding. Overall, 298 respondents returned surveys.

^bPercentage of respondents reporting financial ties to for-profit companies in any sector, including but not limited to the pharmaceutical, device, and biotechnology sectors.

The majority of PAOs that accept industry funding surely strive to not let it influence their decision-making. As Susannah Rose points out, however, “Numerous studies have found that even established and respected researchers and physicians are influenced by drug company money . . . there is no reason to believe that PAOs are any less susceptible to such influence.”^12(p682) Indeed, a recent study found that patient and professional groups with industry ties are more likely to hew to industry-backed positions.¹³

Some may grant that industry funding can bias PAO decision-making but dismiss its significance on the grounds that unlike, say, physician bias, PAO bias carries no risk of serious harm. This is incorrect. First, many PAOs interact directly with patients. If the information conveyed to patients during these interactions is biased, it is concerning for the same reason that biased physician advice is concerning: it can lead patients to make suboptimal decisions regarding the management and treatment of their disease. More concerning are the aggregate effects of bias across PAOs. As a recent commentary suggests, when so many “ostensibly independent players” are subtly biased toward industry interests it can distort the way they think about and develop health policy.³ The subtle but systematic shift of health policy goals toward industry interests and away from patient interests carries serious risk of harm not just to PAOs’ constituents but to the broader public.

BENEFITS

Extending the federal sunshine law to cover PAOs and including PAOs in the OPD would have multiple benefits. First, it would give policymakers, journalists, and others a means of easily assessing the financial ties of PAOs in a way that is not currently possible. Today, when PAOs lobby, deliver testimony, or send comment letters, it is difficult to determine whether and to what extent they have COIs that might influence their positions.^14,15 Consider a recent case in which a coalition of PAOs sent letters to CMS and Congress opposing a Medicare demonstration project testing lower payment rates for Part B drugs.⁸ Neither letter indicated that their signatories had ties to the pharmaceutical industry. Therefore, they gave the impression that the demonstration project, which had been criticized by the pharmaceutical industry, was also widely opposed by grassroots patient organizations. Eventually, a report by Public Citizen revealed that at least 75% of the 147 PAOs that signed one or both letters had received industry money.⁸ Although the study provided critical context for the letters, it required labor-intensive analysis of each of the 147 PAOs’ Web sites and was published months after the letters had had a chance to influence public debate.

Including PAOs in the OPD would provide a real-time means of assessing whether PAOs lobbying for or against particular policies have potentially biasing financial associations. At minimum, such information could be factored into the way that policymakers and the public consider PAOs’ appeals, but it could also be used by journalists, watchdog organizations, and others to direct follow-up inquiries to PAOs. Moreover, with ready access to information about PAOs’ financial ties, government agencies, such as the US Food and Drug Administration, that invite PAOs to participate in policy discussions could use the information to ensure that PAOs invited to talks are not disproportionately associated with industry.

A second benefit of a sunshine law would accrue to PAOs that receive little or no industry funding. With limited transparency, it is difficult for third parties to distinguish PAOs that are industry funded from those that are not. And among industry-funded PAOs, it is difficult to distinguish those that receive modest amounts of industry funding from those that wholly depend on it. These difficulties hurt PAOs that are largely or totally independent from industry. Although policymakers have indicated interest in hearing from independent PAOs, the voices of these groups may be crowded out by industry-backed PAOs, especially because the largest PAOs tend to be industry funded.⁵ Uniform transparency standards would benefit independent PAOs by giving them a way to differentiate themselves from organizations heavily backed by industry.

A third benefit of a sunshine law would also accrue to PAOs. Under the sunshine law, the burden of reporting payments to physicians falls on drug and device companies, which must provide payment information to CMS. If this structure was maintained for PAOs, it would relieve resource-constrained PAOs of the burden of maintaining current donor information on their Web sites. Instead of publishing donor information themselves, PAOs could post a link to the payments database on their Web sites.¹⁶

OBJECTIONS

One possible objection to a sunshine law draws on research highlighting perverse effects of disclosure on the behavior of disclosers.¹⁷ In a series of experiments, researchers found that when an advisor discloses a COI to an advisee, it can actually lead the advisor to offer more biased advice because of “strategic exaggeration” and “moral licensing.”¹⁸ Although this research has been cited in response to calls for greater transparency on the part of PAOs,¹⁷ it is misapplied in this context. Experiments demonstrating the perverse effects of disclosure have involved interpersonal interactions between individuals, which are characterized by different dynamics than are relationships between institutions and the public. There is no evidence to suggest that similar effects would occur when institutions like PAOs disclose their conflicts.

A second possible objection to a sunshine law is that transparency can be achieved via greater self-regulation on the part of PAOs and thus does not require the effort and costs associated with a legislative solution. There are indeed reasons to believe that PAOs are moving toward greater transparency. Recently, the director of the National Health Council (NHC), an umbrella organization that counts PAOs, drug companies, and insurers among its members, announced that all PAO members would soon be required to disclose how much money they receive from medical companies.¹⁹

The NHC should be applauded for its efforts. Nonetheless, even successful pushes for greater self-regulation among PAOs do not obviate the need for a sunshine law. First, it bears noting that there are roughly 50 PAOs in the NHC, whereas there are thousands operating in the United States.^10,20 Even if all members of the NHC and many other PAOs voluntarily became more transparent about their finances, industry-wide, voluntary compliance with uniform transparency standards is unlikely. What’s more, organizations that receive a high proportion of their revenue from industry have greater incentives to avoid transparency. Thus, a possible perverse outcome of voluntary disclosure is that organizations with “less to hide” may actually disclose more than organizations that are heavily backed by industry. From the perspective of third parties, organizations that act responsibly by voluntarily disclosing their finances may appear more conflicted than organizations that disclose nothing, even if the latter have more severe COIs. Legally mandated transparency would avoid these pitfalls by establishing uniform standards across organizations.

A related objection is that even if legally mandated transparency is desirable, it can be established by states and thus does not require a federal law. Recently, Nevada became the first state to pass a law requiring PAOs to disclose funding from pharmaceutical companies.²¹ A similar law has been proposed in Connecticut.²² States’ attention to this issue is laudable, and state laws might usefully complement a federal law. However, there are advantages of extending the federal sunshine law that cannot easily be achieved at the state level. Whereas Nevada’s law applies to PAOs operating in the state—as, presumably, would other state laws—a federal law would apply nationally. Why is this scope important? Consider again the letters criticizing the Part B demonstration. Signatories to letters were based in more than 20 different states. Short of all 50 states passing sunshine laws, the only way to ensure transparency in future cases of this sort is with a federal law. Even with 50 state laws, inconsistencies across laws and the distribution of information across multiple databases would make comparisons among PAOs slower and more burdensome, thereby sacrificing benefits that come from being able to quickly look up and compare organizations.

Additionally, a federal law would have the advantage of building on established foundations. Under the current sunshine law, the leverage for achieving transparency comes not from compelling doctors to disclose payments they receive but from compelling drug and device manufactures whose products are purchased by CMS to disclose certain payments that they make. Thus, even drug and device company payments to doctors who are not reimbursed by CMS must be reported.²³ It is unclear which levers states could pull to mandate transparency. It is possible that other states would follow Nevada’s lead and compel PAOs to disclose their funders as a condition of operating in the state, but this shifts burdens to PAOs. Additionally, the federal sunshine law already establishes penalties for companies’ reporting failures. Rather than developing new penalties at the state level, these same penalties could be carried over in the context of new reporting requirements for payments to PAOs. Lastly, including PAOs in the OPD or another database hosted by CMS could take advantage of technical infrastructure that has already been developed.

A final objection to a sunshine law is that transparency is insufficient: it simply highlights a problem while doing nothing to address it. This objection requires a two-part response. First, even if PAOs do nothing more to address their COIs, transparency alone would have positive effects. As noted, policymakers and others could use the information to solicit input from independent patient organizations, thereby increasing the influence of these organizations while moderating the influence of organizations heavily backed by industry. Moreover, with greater transparency, patients could make informed decisions about the PAOs with which they chose to interact.

These points deserve further elaboration, as this is an area in which the case for mandatory disclosure of industry payments to PAOs is arguably stronger than the case for disclosure of industry payments to physicians. The underlying rationale for the current sunshine law is that with greater access to information about physicians’ COIs, patients can make an informed choice about whether to see a doctor who receives industry payments. However, there are significant obstacles to patient utilization of Open Payments data. First, evidence suggests that few patients know of the existence of the OPD and fewer still use it.¹⁶ More problematically, even if the number of patients accessing the OPD eventually rises, many patients lack a realistic choice between doctors, and others cannot afford the opportunity costs involved in searching for a new doctor.²⁴

Neither obstacle would prove as daunting for a PAO payments database. To begin with, whereas the primary intended users of physicians’ payment data are patients, the users of PAOs’ payment data would include journalists, watchdog groups, and policymakers who have already shown an interest in knowing more about the COIs of PAOs and who would be prepared to use a database. More importantly, whereas patients need to see doctors, members of the public need not engage with particular PAOs. If a patient or policymaker is concerned about the COIs of a PAO, she can choose to question, criticize, or simply ignore its advice or policy proposals. Alternatively, she can seek out information from another PAO more easily than most patients can find another doctor.

Ultimately, then, even if transparency does not alter PAOs’ behavior, it can change the way that the public engages with them. However, the second point to note is that a sunshine law would likely influence PAO behavior. If PAOs have to disclose their funding, some may choose to accept less industry money. Others may stop accepting industry money altogether. Some PAOs would chose to maintain relationships with industry. But bringing these relationships into the sunlight may elicit positive responses from PAOs. It may prompt PAOs to strengthen existing COI policies. Additionally, it may prompt PAOs to state more explicitly why they believe the benefits of financial associations with industry outweigh their risks. Initiating this kind of open conversation about industry–PAO relations would be an important step toward increasing PAOs’ accountability to their constituents and the broader public.