Introduction

More than two years after the historic passage of the Patient Protection and Affordable Care Act (ACA), ongoing debate over US health care reform has reinforced how profoundly politics can shape health policy. The results of the 2012 Presidential election will have a profound impact on the delivery of health care in the US for years to come [1].

Specific elements of health policy, however, are traditionally viewed as being relatively insulated from political influences. For example, insurers and health care systems in many countries provide formulary coverage for prescription medications and must balance rising costs against appropriate access to new treatments [2]–[4]. A number of agencies help to determine whether new drugs should be listed in drug formularies, and the principles that drive their decisions (such as evidence of effectiveness and safety, evidence of need, and cost implications) have been reviewed [2], [3], [5]. Although cost-effectiveness criteria are used to guide policy related to drug coverage in many countries, drug reimbursement decisions within publicly funded health care settings in the US largely exclude considerations of cost [4]. For example, the ACA specifically prohibits use of cost effectiveness thresholds to guide coverage decisions [6]. The emergence of very expensive treatments, such as biological treatments for cancer, has highlighted the issue of cost in making coverage decisions [7], [8].

The Canada Health Act supports a near-universal system of health care across the country's ten provinces, although each province makes its own decisions about which prescription drugs it will cover (Appendix S1). Despite well-established principles to guide decisions, significant differences have been observed in formulary coverage of drugs across Canada, with the timing of drug additions onto different provincial formularies varying in some cases by years [9], [10].

External forces may help to explain these variations. For example, some have speculated that undue political pressures may have influenced certain drug formulary decisions [11], [12]; however, this theory has not been formally tested. Capturing the many nuances of political influence can be challenging, but elections are easily quantified and represent a particularly acute form of political pressure. To assess the relationship between funding decisions and this form of political pressure, we examined the association between the timing of provincial funding announcements for a class of drugs known as the cholinesterase inhibitors and the timing of elections in these provinces.

Alzheimer's disease is the sixth leading cause of death in the US, and there are still no effective treatments to prevent, halt or reverse this condition [13], [14]. Cholinesterase inhibitors were the first drug treatments approved for Alzheimer's disease and related dementias. Health Canada approved donepezil (Aricept) in 1997, rivastigmine (Exelon) in 2000, and galantamine (Razadyne, Reminyl) in 2001. Cholinesterase inhibitor use is widespread, with global sales of donepezil reaching $4.4 billion dollars in 2010 [15]–[18]. In Ontario, cholinesterase inhibitor prescriptions grew dramatically between 2000 and 2011 (Figure S1).

We chose to focus on cholinesterase inhibitors for this case study because they represented the first major therapeutic advance for dementia. Several debates about their clinical and cost effectiveness arose between the premarketing stage and the recent arrival of generic formulations signaling the final stage of their product life cycle [17]–[19]. A recent trial confirms the clinically marginal benefits of continued cholinesterase inhibitor treatment in patients with more advanced dementia [20], and there is now general consensus that these drugs possess modest efficacy [17]–[19].

Methods

We reviewed reports that covered the regulatory approval of cholinesterase inhibitors by Health Canada, the efforts of pharmaceutical manufacturers and patient advocacy groups to have these drugs reimbursed by provincial drug formularies, and announcements from provincial governments about formulary coverage for these medications. Canada's approach to drug approval and reimbursement is described in Appendix S1. We did not examine drug formulary decision-making or elections in Canada's three territories (Nunavut, the Northwest Territories, and Yukon) because of their relatively small population and their unique demographic profile. Much of the population of these territories is First Nations and they receive prescription medications from special federal and territorial programs. We performed a comprehensive search of public agencies, media outlets, and Google News, to identify news media reports with the dates of important funding announcements, and the PubMed database to identify relevant medical journal articles. We then reviewed proceedings from provincial legislatures that included transcripts of discussions about public reimbursement for the cholinesterase inhibitors. Whenever possible, we determined the earliest date that funding decisions were announced. We identified dates of provincial elections from official elections websites for each province (e.g., http://www.electionsmanitoba.ca/).

Results

Cholinesterase inhibitors were added to formularies in all ten Canadian provinces between June 1999 and October 2007. None of these decisions were modified or reversed when followed up to September 2012. Four of the ten drug funding announcements occurred within 60-day intervals preceding provincial elections (Table 1). This observed proportion was significantly greater than that expected by chance (P = 0.032).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Timeline detailing announcements of cholinesterase inhibitor inclusion onto provincial formularies and dates of provincial elections^a.

https://doi.org/10.1371/journal.pone.0056921.t001

The four affected provinces were Ontario, Manitoba, Nova Scotia, and Newfoundland and Labrador. Detail regarding the chronology and context of events for each of these provinces is provided below and in Figures S2, S3, and S4.

Discussion

Canada's ten provinces announced coverage for cholinesterase inhibitors over a nine-year period stretching from 1999 to 2007. This marked variation likely reflects the influence of various factors including uncertainty about the evidence base supporting the use of these drugs, media reporting, lobbying efforts by pharmaceutical manufacturers and patient advocacy groups, and competing priorities that were unique to each province. The complex interplay of these multiple influences shifts continuously during the life cycle of any drug, from its initial introduction onto the market with limited information from RCTs, through growing clinical experience and postmarketing surveillance, until well after the availability of generic formulations of the product [40]. Beyond this dynamic interaction of multiple influences, we found evidence in this example of a relationship between drug funding announcements and political pressure in the form of impending elections. In our case study, four of ten provinces added cholinesterase inhibitors to their formularies in the 60-day period preceding an election, including the first two provinces to list these drugs. At least two politicians stated that the timing of announcements was influenced by political pressures [31], [35].

These findings provide an additional explanation for the variability in drug coverage observed in different jurisdictions [9], [10]. Qualities specific to individual provinces must also be considered: for example, the size of the population likely to require the drug, and the funds available for health care in a particular budget cycle [5], [41], [42].

The case of the cholinesterase inhibitors also exemplifies the challenges involved in deciding when sufficient evidence has accumulated to warrant formulary coverage for expensive new drug treatments [7], [8], [43]. Two provinces initially cited the disappointing results of the AD2000 trial as a reason for rejecting formulary coverage for these drugs [37], [44]; subsequent decisions in both provinces to fund the drugs supports the notion that influences other than scientific evidence may trigger drug policy decisions.

Our focus on cholinesterase inhibitors strengthened this study for several reasons. First, this drug class included the first approved drug treatments for Alzheimer's disease, a common and devastating illness. There would be less interest and political gain in the approval of a drug that was simply an addition to an existing and previously populated drug class (i.e., a “me too” drug) or one that targeted a disease with a less fearsome public profile. Second, the cholinesterase inhibitors have nearly completed the drug life cycle from pre-approval to the recent arrival of generic formulations. This allowed us to take a long-term view and thoroughly analyze the funding decisions that were made along the way. By comparison, examination of newly approved drugs is often difficult given uncertainty about their long-term effectiveness and safety. Drug formulary decisions are usually deferred until such evidence becomes available. Finally, the cholinesterase inhibitors are still relevant, as they remain the most commonly prescribed medications for Alzheimer's disease and related dementias. Recent publications about these drugs include a commentary discussing the strategies undertaken to extend the patent life of donepezil [45], and a RCT called DOMINO that demonstrated marginal benefits associated with continuing donepezil in people with moderate to severe dementia [20].

Other literature about the relationship between political forces and health policy decisions merits consideration. Some reports have speculated about the link between elections and the timing of drug funding announcements, but have not attempted to quantify this association [46]. Examples involving the human papillomavirus (HPV) vaccine Gardasil [11], [47] and the adjuvant breast cancer treatment trastuzumab (Herceptin) [12] demonstrate that drug policy decision-making requires balancing evidence of clinical effectiveness, safety, and cost-effectiveness on the one hand, and external factors – such as media reports, lobbying by various parties, and political interests – on the other. Additional factors such as the need for legislators to make decisions quickly despite limited expertise may also affect the timing of decisions [48].

The link we show here between upcoming elections and drug funding announcements raises concern because the external pressures and stringent deadlines involved in elections may adversely affect health policy decisions. To illustrate this principle, consider the drug review deadlines established at the Food and Drug Administration (FDA) after the Prescription Drug User Fee Act (PDUFA) was enacted. As compared with drugs approved at other times, drugs approved in the two months before their PDUFA deadlines were more likely to be withdrawn for safety reasons, more likely to carry a subsequent black-box warning, and more likely to have one or more dosage forms voluntarily discontinued by the manufacturer [49].

Politics has a necessary and fundamental role in shaping health policy decisions [50], but safeguards are needed to prevent undue political maneuvering by self-interested stakeholders. An example of the potential consequences of political interference on healthcare legislation involves the introduction of Medicare Part D, the US program that subsidizes the costs of prescription drugs for seniors and the disabled [51]. Part D was enacted as a component of the Medicare Modernization Act of 2003. Some members of Congress initially proposed authorizing the federal government to negotiate for lower drug prices. However, Congressional leaders (some of whom later left their government jobs to become lobbyists employed by the pharmaceutical industry [52]) included a clause in the final bill that prohibited the government from negotiating with pharmaceutical companies. In contrast, government is allowed to negotiate drug prices for Medicaid and programs sponsored by the Department of Veterans Affairs, allowing these programs to pay less for drugs than Medicare. Estimates suggest that extension of existing price setting mechanisms to Medicare Part D could result in annual savings of over $20 billion [53].

Our study has important limitations. First, it is based on evaluation of a single drug class, and it is certainly not generalizable to all medications. However, we have argued above for the value of a case study approach and for our focus on the cholinesterase inhibitors. Second, we had limited access to information on the impact of lobbying efforts by manufacturers and patient advocacy groups on governments as they made funding decisions. Negotiations between manufacturers and government can sometimes lead to incentives (e.g., cost sharing) that help to shape the final funding decision. Third, it is important to distinguish the timing of a public announcement of drug funding from the earlier time that the funding decision was reached. Government officials in Canada typically make such decisions after considering reimbursement recommendations from a formulary review committee, but information about the timing and content of recommendations made by formulary committees and the subsequent decisions reached by government are often not made public. We only had information on the timing of public announcements. In this context, a strong case can be made for increased transparency at all steps of the drug approval and reimbursement process [54].

Supporting Information