By Jenny Manzer – 2005 Michener-Deacon Fellowship recipient (published December 2006)
Gulping a daily dose of statins to tame your cholesterol might lead to amnesia. Taking a pill to cope with impotence could cause sudden vision loss. Hormone therapy probably won’t help a woman’s heart, but it might increase her risk of incontinence.
These are just some of the unexpected side-effects of popular drugs that only came to light after they were on the market and being taken by millions of people. While some of these effects are rare, adverse drug reactions are as common as city smog. More than one in three Canadians report having personally experienced an adverse drug reaction, according to a 2003 Health Canada survey.
Estimates suggest half of all new drugs on the market end up having side-effects that aren’t seen in the initial testing phase – and many of them are serious. A 2002 U.S. study, published in the Journal of the American Medical Association, estimated that one in five new drugs will eventually receive a “black box” warning or be pulled from the market altogether. No one knows for sure how many Canadians die from drug reactions – studies have suggested anywhere from 1,400 up to 10,000 people a year – but it’s clearly a major health problem.
The clinical trials that test our medications are usually shorter than real-life treatment and done in a select group of “healthy” patients, so we must do a good job of tracking how drugs perform in the real world. Unfortunately, by most accounts, we don’t. Critics say a long history of under-funding has made post-marketing surveillance the weakest arm of Canada’s drug safety system.
While there’s agreement that post-market monitoring needs an overhaul, there are differing ideas about how to do it. Dr. Geoffrey Anderson likens our post-market surveillance system to a stool that needs three sturdy legs to function. In his analogy, one leg tracks adverse drug reactions. The second does randomized controlled trials after the drug is on the market, which involves assigning a group of patients to receive the medication then comparing them to another group, who receive a placebo. The third leg performs observational studies, which simply follow patients who are already taking the drug to see how they fare.
Dr. Anderson, a professor of Health Policy, Management and Evaluation at the University of Toronto, says to date, Canada has tended to focus post-market monitoring on adverse drug reporting. “We seem to be putting a lot of investment into adverse drug reporting, which is one very important part of the process and certainly a good way to pick out rare, very serious idiosyncratic side-effects,” he says, “but it doesn’t do a very good job of identifying adverse effects that are more common. I guess a classic example would be Vioxx and heart attacks.”
Other experts say Health Canada should keep drugs on a tighter leash once they’re on the market. One way to do this would be through conditional licensing. A drug would only be approved if the manufacturer agrees to a certain set of conditions, such as promising to do large, follow-up studies of its real-world use. That way, the drug is still available promptly, the public gets information about how it performs and Health Canada has a mechanism to remove the drug if problems emerge.
Conditional, or probationary licensing, has its cheerleaders. The House of Commons Standing Committee on Health, which issued a report on prescription drugs in 2004, suggested our post-marketing surveillance system needs major changes. The committee recommended that Health Canada make licensing of new drugs probationary to ensure that post-marketing surveillance is carried out diligently after the drug is approved.
As it stands, Health Canada rarely issues a drug approval – called a Notice of Compliance – with conditions. In fact, there were only 27 Notice of Compliance with Conditions issued from 1998 to 2005, according to figures from an Access to Information request. Most of the conditions given in those 27 approvals have yet to be fulfilled. As of September 2005, only nine of these conditions had been completed, according to documents from an Access to Information request.
In practice, Health Canada is faced with a bit of a conundrum, says Dr. Anderson. Regulators must deem a drug to be safe or not safe, so these Notice of Compliance with Conditions are usually only given in cases where the benefit of the drug could be big – which is not the case with most new medications, he adds. Typically these types of approvals are used for treatments involving life-threatening conditions such as AIDS or cancer.
Dr. Supriya Sharma, associate director general of Health Canada’s Therapeutic Products Directorate (TPD), says the Notice of Compliance with Conditions is given in cases where a drug shows promising efficacy for a serious condition, not when there are safety concerns. But Dr. Sharma says the TPD is now studying the entire regulatory framework and weighing whether it needs different gradations of licensing to provide more leverage. “We’re really looking at it to see what’s the best way to do ongoing evaluation. I think in the previous system, we really thought about on the market, off the market, but this is focusing on the continuum.”
One problem with conditional licensing is that drug companies have a poor track record of following through with the promised trials. A 2000 study by the U.S. lobbying group Public Citizen looked at all the new drugs the U.S. Food and Drug Administration (FDA) approved from 1990 to 1999 that required some kind of post-market commitment from the company.
Public Citizen found that only 13% of the study commitments made for drugs approved from 1990 to 1994 had been fulfilled. None of the promised studies for drugs approved from 1995 to 1999 had been done. Worse, Public Citizen found the FDA has no power to ensure these promises are kept. The finding was confirmed by the FDA’s own researchers last year, who found that manufacturer compliance in fulfilling post-market studies was poor, with results overdue years after they were required.
The idea that both the U.S. and the Canadian systems lack the teeth to compel drug companies to complete required studies leads some experts to support another school of thought: gaining leverage at the formulary level, when we decide if our public plans will pay for it.
The drug enters the market, but will be only be added to the public formulary – or list of medications our healthcare system will fund – under certain terms and conditions. For example, the public payers could call for a well-designed observational study or a large randomized trial once the drug is on the market. These types of studies could be done by the drug company, or by the public sector, as was the case with the Women’s Health Initiative, the large U.S. study that revealed the risks of hormone therapy in 2002.
Boosters of this approach, such as Dr. Anderson, say that trying to change the system at the licensing level would be a logistical migraine, especially since it would draw in legal baggage from other countries because of various rules and statutes. A better, made-in-Canada solution, he insists, would be to reassess the drugs once they are being considered for the public formulary, or even for inclusion into private insurer plans. The provinces could give a drug a provisional listing, he argues. If the drug didn’t perform well in the real world, the provinces could simply stop paying for it. “So my sense is that the real area where we have traction is not the federal level, but rather at the provincial level,” he says.
Dr. Anne Holbrook, director of clinical pharmacology at McMaster University in Hamilton, agrees that changing the system at the licensing level would be odious. Given the influx of expensive medications on the market, she can envision a system in which approval by Health Canada is only the first step in the public reckoning of a new drug. If a company wanted the drug included on the formulary, it would have to go through more hoops.
Health Canada doesn’t seem to have much power once a drug is on the market, but there is hope for the formulary, says Dr. Holbrook, who sits on the Canadian Expert Drug Advisory Committee, an independent board that makes recommendations to publicly funded drug plans. “The way things are going at least, the provinces and the feds are beginning to work more collaboratively to have a kind of a common front.”
Dr. Holbrook says researchers had hoped that the Notice of Compliance with Conditions would be a useful tool to retain some control over drugs on the market and offset faster approval times. Unfortunately, this kind of conditional licensing hasn’t panned out the way many of her colleagues had expected, she says.
Experts say inadequate funding for post-market surveillance is part of the problem. While the problem of slow drug approval times has received a lot of attention from Health Canada – and a big chunk of cash – post-market monitoring is still sitting in coach class by comparison.
In 2006, the budget for the Therapeutic Products Directorate, which handles new drug submissions, was $42 million a year. The Marketed Health Products Directorate (MHPD), which monitors the safety of drugs on the market, received only $13 million a year. The MHPD receives no industry funding, while the TPD gets more than half of its budget from industry user fees.
Mary Wiktorowicz, chair of York University’s School of Health Policy and Management, says chronic under-funding over the years has led Health Canada to its current tight relationship with industry.
“I think that’s part of the weakness with Health Canada and why they tend to rely on industry so much, because they haven’t had the resources. They’ve had to find other ways to work with industry, get industry to comply with their standards because they are under-resourced.”
Colleen Fuller, co-founder of the Vancouver-based consumer group PharmaWatch, says post-market monitoring has long been hungry for cash. “Within Health Canada, within the whole area of Therapeutic Products, it is the poor cousin within the entire chain,” says Fuller.
“From my point of view, it’s basically because there is no interest in it from the drug industry itself and the job of Health Canada is to secure the investment environment. This is what it boils down to. And if the drug industry isn’t interested, then nor is Health Canada.”
Still, both the budget and staff at the MHPD are growing, as its director, Dr. Marc Berthiaume, is quick to emphasize. In 2006, his department had about 120 staff, up from just 63 in 2003. Although the directorate had only 15 people reviewing adverse drug reactions for pharmaceuticals, that number is also on the rise. Asked if 15 are enough, Dr. Berthiaume says they’re moving in the right direction.
“I don’t think we could grow faster than that. You need to train these people. You need to develop the work process around them, and pharmacovigilance or regulatory medicine is not something that you can have at university as a course, so basically part of it is training at work.”
The 2006 numbers seem scrawny, however, compared to the 182 assessment officers who worked on drug reviews at the TPD, which had a total staff of about 525 people.
In the 2004 report, the Standing Committee on Health expressed concern over the lack of staff to address adverse drug reaction reports and called for additional resources to assess reports and convey the findings back to consumers.
Dr. Joel Lexchin says we have to make it clear to the public that reporting adverse drug reactions is important, and to make it easier to do so. We also have to give feedback to patients who file reports. “You have to be seen to be doing something with the information that you’re getting, so people don’t feel it’s disappearing down a black hole.”
It’s not enough to sit and wait for good or bad news, insists Dr. Lexchin, a professor at York University’s School of Health Policy and Management. He says we need to set up a system of active monitoring for certain drugs, particularly those used by high-risk groups. “So you can set up a database whereby you record the names and birth dates of, say, the first 50,000 people who get a drug and you can link with things like hospital visits or hospitalizations or death or cancer or doctor’s visits using other databases, and you can see whether or not these drugs are presenting certain problems.”
Drug safety experts seem to agree that linked databases are the way to go. Canada has a rich vein of data sources to mine, thanks to our public healthcare system. In B.C., for example, every time a prescription is filled, an electronic record is created that leaves the patients anonymous, but tells us what drug they took and in what dose. The information can then be linked to other data such as hospitalization records, physician visits or mortality to determine how people who took a given drug fared compared to those who did not.
“I think Health Canada’s approach has been very narrow,” comments Mary Wiktorowicz. Basic science has changed in areas such as biotech products and so has health services research, she points out. We now have access to databases that weren’t available a decade ago. “So in terms of being able to monitor products when they’re on the market, we have some powerful resources and a powerful research community that is available to do that kind of monitoring that is not being used.”
“Canada could be a leader in this area in terms of innovative post-marketing surveillance,” she says, a refrain echoed by many researchers.
These databases are being used sporadically. For example, the team at the Institute for Clinical Evaluative Sciences (ICES) in Toronto has done several studies of elderly patients using Ontario drug plan data – but the information could be used to far greater advantage. Dr. Anderson says the FDA tends to work strategically with university-based researchers on observational studies and it also has a large in-house scientific team, both of which Canada lacks.
Dr. Andreas Laupacis and colleagues from ICES have floated a new approach to drug evaluation in Canada. Their idea, outlined in a 2003 Canadian Medical Association Journal (CMAJ) article, places more emphasis on using the drug’s performance in the marketplace to decide if provincial governments should add it to the public drug plan. The group wants to see “centres of excellence in pharmacosurveillance” set up across the country. Large, head-to-head randomized trials could be done after the drug has been approved, the team says. This will help determine the “relative benefits and risks of competing medications in regular practice,” they write. These centres of excellence could provide a badly needed network to share data between provinces and detect trends in side-effects.
While we must do a better job tracking drugs, we also need to boost adverse drug reaction reporting. By Health Canada’s own estimates, as few as 10% of side-effects are currently being reported, though other estimates suggest it may be as low as one percent.
Various parties have supported the idea of making adverse drug reaction reporting mandatory, including former Health Minister Ujjal Dosanjh and the 2001 coroner’s jury that made recommendations following the death of teenager Vanessa Young, who had been taking the stomach drug Prepulsid.
At the moment, Health Canada is still consulting on mandatory reporting, according to Dr. Marc Berthiaume. Some drug safety advocates are outraged that adverse reaction reporting is still voluntary. Physicians are less keen on the idea. A Health Canada survey showed that 45% of doctors oppose mandatory reporting. Clinicians may regard reporting as a hassle, but some researchers also doubt mandatory reporting will result in a better system.
Critics of mandatory reporting say it will be easy for physicians to dismiss patient symptoms as not being drug-related. They also question how all the data that flood into Health Canada will be analyzed to produce meaningful information.
Dr. Andreas Laupacis of ICES says he suspects such a system would result in an avalanche of minor adverse drug reaction reports. He hasn’t filed a single report in his more than 20 years of practice. He’s not alone. Only a small fraction of health professionals say they have reported an adverse drug reaction in the past year – and more than a third of doctors say they don’t even know how to file one.
While Dr. Lexchin also doubts mandatory reporting will be a big improvement, he stands by the importance of having the public make reports. In a commentary in the CMAJ, he argues that these kinds of reports capture problems, trends and nuances that can be missed by only relying on formal studies, and suggests several ways to improve the reporting system.
One interesting idea he notes is the British example of flagging new drugs with black inverted triangles. The triangle symbol appears beside the drug name in the formulary and reinforces the need for physicians to report any suspected adverse reactions. In the U.K., these black triangle drugs undergo close monitoring for two years, and the symbol stays until safety is established. This system could be adapted for Canada, Dr. Lexchin argues, with symbols appearing in our listings of drugs.
While adverse drug reaction reporting is still voluntary for health professionals, drug companies must meet certain post-market obligations under the Food and Drug Regulations. For example, companies must forward any adverse drug reaction reports they receive to Health Canada within 15 days. In 2004, Health Canada started doing inspections of Canada’s health product manufacturers and distributors to determine how well companies were complying with post-market obligations. Health Canada completed 45 inspections from September 2004 to August 2005, then another 18 from September 2005 to December 2005. The Health Canada department that does these inspections draws more than 60% of its budget from user fees paid by drug companies.
All of the businesses, which included both over-the-counter and prescription drug companies, received a rating of “C.” This grade indicated that they were found to be compliant with their regulatory requirements with no observed “objectionable practices.”
Still, more than half of the companies who underwent these initial inspections had “observations” noted which required some kind of corrective measure. There were 35 observations noted, of which 12 were “major” – one level down from the most serious rating of “critical.”
A company could receive a “major” rating for not reporting adverse drug reactions within the 15-day time limit, or for not turning their records of unusual failure of new drugs over to Health Canada within 72 hours.
There are new initiatives underway, such as these post-market inspections, and other innovative ideas in waiting in the wings. In November 2006, Health Canada announced that it has awarded a contract to supply a new computerized system for monitoring adverse reactions to health products. Health Canada says the new system, expected to be up and running in October 2007, will bolster their ability to detect safety problems and allow for better analysis, such as being able to track products over their entire life-cycle. Still, advocates fear progress on post-market surveillance will fizzle.
Several measures brought forward under the Liberal regime in the wake of the Vioxx media flare have been delayed with the change of government, including mandatory reporting and the creation of a Health Products Safety Board. The board, announced in February of 2005, was intended to ensure transparency and encourage public input into regulatory decision-making. A post-market surveillance strategy was supposed to be finalized and implemented in 2005, but that has also yet to happen, although there’s a draft document on the go, according to Dr. Berthiaume.
Some drug safety advocates fear the political will to change the system is gone – at least until the next major crisis. Unfortunately, many experts also agree, there will be one. It’s just a question of when.
“If you look at the history of adverse drug report monitoring in Canada, it’s all been crisis-driven. It was set up in the shadow of thalidomide,” reflects Colleen Fuller from PharmaWatch, referring to the infamous drug that caused severe birth defects. It was pulled from the Canadian market in 1962. “Nothing’s changed as far as I can tell,” she says. “So right now the only thing that will get it going is another crisis.”