Proceedings of the Standing Senate Committee on
Social Affairs, Science and Technology
Issue 25 - Evidence - November 8, 2012
OTTAWA, Thursday, November 8, 2012
The Standing Senate Committee on Social Affairs, Science and Technology met this day at 10:29 a.m. to study prescription pharmaceuticals in Canada (topic: post approval monitoring).
Senator Kelvin Kenneth Ogilvie (Chair) in the chair.
[English]
The Chair: Honourable senators, I welcome you to this meeting of the Standing Senate Committee on Social Affairs, Science and Technology. I will acknowledge our witnesses specifically when we introduce them to request their contribution, but it is already obvious to the committee that one of our witnesses today is joining us by video conference. I urge you all to speak directly into the microphones when you are asking questions so that the transmission is clear through the lines to our guest who is online.
I am Kelvin Ogilvie, a senator from Nova Scotia and chair of the committee. I will ask my colleagues on the committee to introduce themselves.
Senator Seidman: I am Judith Seidman from Montreal, Quebec.
Senator Martin: I am Yonah Martin from British Columbia. Welcome.
Senator Verner: I am Josée Verner from Quebec.
Senator Demers: I am Jacques Demers from Quebec.
Senator Eaton: I am Nicole Eaton from Toronto.
Senator Cordy: I am Jane Cordy from Nova Scotia.
Senator Dyck: I am Lillian Dyck from Saskatoon.
Senator McInnis: I am Tom McInnis from Nova Scotia.
Senator Eggleton: I am Art Eggleton from Toronto, deputy chair of the committee.
The Chair: Thank you, colleagues. I will recognize our witnesses in the order in which they appear on my list. Mr. Lemmens, the honour goes to you this morning to begin the presentations to us.
Senators, Professor Lemmens is coming to us today from Oxford.
Trudo Lemmens, Scholl Chair in Health Law and Policy, Faculty of Law, University of Toronto, as an individual: Thank you very much.
[Translation]
First, I would like to apologize to the French-speaking members of the committee for not having had my brief translated.
[English]
I submitted it too late for translation, but I understand that it will be translated in the coming days. It is a memo written by myself and my research associate, Shannon Gibson.
I will start with an introductory comment about the historical context in which this discussion of today occurs. Drug regulations were first introduced to remove toxic medicinal products from the market. Although a basic efficacy requirement was added in the 1970s, it has not led to a sophisticated review of the merits of new drugs. In fact, the gradually increased requirements of submitting safety and efficacy data from a few core clinical trials have gradually led to the development of a clinical trials industry which people have to realize focuses primarily on the fulfillment of an administrative requirement. It does not have the mandate to analyze whether this production of pre-approval data gives us meaningful and sufficient information. This is not what the industry is asked to do and it is not what it has a real interest in doing.
Unfortunately, many people, including physicians, ignore the limitation of premarket data which are produced under the control of those who have a vested interest in getting the drugs quickly on the market and not necessarily an interest in detecting problems. The system has, to some extent, a perverse effect. It creates a false assurance of safety and effectiveness which, in the current context, has sometimes been used as a marketing tool without clear control on how drugs are used in the real world.
I want to discuss briefly the following issues with respect to post-marketing surveillance: First, the lack of good evidence; second, the lack of comparative evidence on patient benefit between different drugs; third, the phenomenon of off-label prescribing, which I understand the committee has already discussed, but which must be associated with post-marketing surveillance; fourth, the need for improved adverse drug reporting; and, fifth, the introduction of post-marketing requirements. I will be brief on all of these issues.
Clinical trials tend to miss important reactions that only occur infrequently or in the long run. They are conducted under conditions that do not reflect real world use of drugs. Patients are exposed to other drug products or underlying diseases and are usually older, sicker and less supervised than clinical trial participants. This creates more serious problems in the context of the blockbuster drug model which is more and more used to treat chronic diseases and also as preventive medicine. When drugs are massively prescribed, even a minor undetected risk can result in thousands of adverse reactions and deaths, as has occurred in some controversial cases.
Another issue is the control of industry over clinical drug trials pre-market, which has been identified as often being biased because of sometimes crafted research design, choices made in the context of statistical analysis, problems with reporting, and other issues. Yet, Health Canada largely relies on the data produced in this context and it does not control how data are presented in medical literature, which is really what guides prescription behaviour of physicians, but we know that there are serious problems with the reliability of the reporting in the medical literature.
The challenge is augmented in the context of so-called personalized medicine. Drugs to treat rare diseases or for niche markets, such a pharmacogenomic products, are often fast-tracked during the drug approval process or provided with more lenient realistic data requirements. The limited approval basis makes it even more critical to assess, after approval, the benefits and risks of these drugs through post-marketing studies. If these are not, or not properly, conducted, potentially ineffective and unsafe drugs can remain on the market for a very long time.
On the second issue, the lack of comparative effectiveness information, it should be clear that the comparative data that is available, which is limited on drug products, often suffers from the same problems of bias that I already mentioned. Unfortunately, in Canada inferior drugs can enter the market and can remain there unless a product is found to be unsafe. Yet, the subtle promotion of inferior drugs, which is possible, can clearly harm patients.
The third thing I want to mention briefly is issues around off-label prescribing. As the committee knows, this is legal and understandable to some degree, but many physicians are unaware when they are prescribing drugs off label. A certain amount of off-label prescribing can be defended, but the exceptions should be tightly controlled.
Unfortunately, companies do not always have an incentive to obtain reliable data on safety and efficacy of off-label prescriptions. Trials are costly and, if they end up showing lack of efficacy, they can destroy an important market segment for their drug, so in case of doubt the company can play it safe and simply allow off-label prescription to continue.
I will invite the committee to look at a recent development in France where recommendations for a new system have been introduced which provides an interesting way to supervise, at least to some degree, the prescription of drug for unapproved uses.
The fourth issue is adverse event reporting. In Canada, the most common means of adverse drug reaction monitoring is through voluntary reporting by health professionals and consumers. Mandatory physician reporting of adverse drug reactions could strengthen post-market surveillance, but reporting is time and labour intensive. Remunerating physicians to report adverse drug reactions could provide a good incentive to substantially increase the amount and the quality of the reports.
I would also urge the committee to look at the system for collecting, storing and analyzing ADR data that is currently in place in the context of Health Canada. Its system, MedEffect, has been described as poorly structured and as compiling information that is often very difficult to interpret. There is a need for increased transparency from Health Canada on how ADR reports are used to inform the post-market surveillance process.
I will conclude by saying something on the issue of progressive licensing. Canada has been working since 2005, which is basically seven years, on a progressive licensing framework that would give regulatory powers to require ongoing assessment of drugs. Unfortunately, the proposed framework still relies primarily on drug manufacturers to conduct their own post-marketing studies.
In my view, there is sufficient evidence from past controversies that raise doubts about how reliable these will be. The ideal solution would be to remove industry control over the conduct of all clinical trials and improve government oversight by, for example, simply removing the direct relation between the clinical trials industry and the pharmaceutical industry. This more drastic reform may not be for tomorrow. It is therefore worth mentioning one positive initiative in this area which could and should be strengthened, which is the Drug Safety and Effectiveness Network.
DSEN relies on post-marketing research conducted through a virtual network of Canadian centres for excellence but there are, notwithstanding its interesting components, two very important caveats about the DSEN.
First, its budget pales in comparison to the marketing and clinical trials budgets of large pharmaceutical companies and, second, DSEN's independence is weakened by the fact it is housed currently under the Canadian Institutes of Health Research, which itself is now promoting closer collaboration with industry, particularly in the organization and conduct of these clinical trials. It is also worth mentioning that CIHR recently appointed two pharmaceutical industry executives to its boards and there has been debate about how this undermines the independence of the agency itself.
My conclusion is that reform efforts need to focus on moving toward better measurements and assessment of drugs, based on clinical outcomes and on explicit and stringent requirements to conduct post-marketing clinical research, to obtain reliable evidence of the effects of drugs in real world settings.
The government should first promote or even require comparative efficacy studies; second, should improve post-marketing adverse event reporting; third, provide regulatory authority to Health Canada to require post-marketing continuing review requirements; and fourth, should promote independent clinical trials data through a change in the structure of the relations between industry and those conducting clinical trials.
The Chair: Thank you, Mr. Lemmens.
I will now turn to Françoise Baylis, Professor and Canada Research Chair, Faculty of Medicine, Dalhousie University. Welcome.
Françoise Baylis, Professor and Canada Research Chair, Faculty of Medicine, Dalhousie University, as an individual: Thank you for inviting me to speak with you again today.
[Translation]
It is always a pleasure to have the opportunity to be here in Ottawa. I will be happy to answer your questions in French. I believe you have the texts in both languages, but I am going to make my presentation in English.
[English]
My presentation today will focus on post-marketing trials involving women, pregnant women and lactating women, but first an important caveat: I agree with many, including most recently Thomas Moore from the Institute of Safe Medication Practice in the United States, that "getting faster access to newly developed, less thoroughly tested drugs is at best a mixed blessing." That is to say, I have serious reservations about the move to get drugs to market sooner at the expense of fuller and more robust information about safety and efficacy.
I commend to you a recent study from October of this year by Joel Lexchin, showing nearly a third, 34.2 per cent, of the drugs that got from fast-track priority review and approval from Health Canada between 1995 and 2010 got a safety warning or had to be withdrawn for safety reasons. That is not an insignificant percentage.
Having said that as a caveat, I am here today to argue for important safeguards for women, pregnant women, and women who are breastfeeding. I will conclude with a series of specific recommendations.
I will skip the definitions I provided of the Phase I to Phase IV trials. Senators can read that later. However, in principle, it is important to emphasize that the goal of Phase IV post-marketing trials is to address evidentiary gaps in comparative effectiveness, drug safety and real world utility. Why are we doing this? We are doing this in pursuit of social benefit. In practice, however, it often appears that these trials are co-opted for private benefit.
Common criticisms of current practice surrounding post-marketing trials include the following: Co-optation of the research system to "seed" off-label use of new drugs, devices and biologics; production of biased evidence as a result of statistical under-powering, absence of comparator arms, withholding of adverse events and altering primary end points; and, last of all, publication bias.
I will touch briefly on some of these issues but, again, with specific reference to the populations that I am concerned with, starting with sex-based biology, understood as "the study of biological and physiological differences between men and women." In my view, post-marketing research needs to look for potential sex-related differences in the safety of efficacy of drugs, devices and biologics, and that includes vaccines and, more specifically, needs to actually track and analyze the outcome for women, pregnant women, women who are breastfeeding and their offspring.
My presentation today takes as its starting point the draft Health Canada Guidance Document on Considerations for inclusion of women in clinical trials and analysis of data by sex, released for public comment in January of this year. I am not sure of the status of this document and where it will move, but it is important to say that my comments fit with that position that Health Canada has taken.
The Health Canada Guidance Document encourages the generation and consideration of new scientific knowledge about sex differences and about therapeutic products used during pregnancy and while breastfeeding. It also recognizes the importance of building the evidence base throughout the life cycle, and that includes post-marketing trials. I will speak briefly to each of the populations just to give you a sense of the concerns I have.
First is therapeutic products used in women. Adverse events reporting and the conduct of well-designed, post-marketing safety clinical trials are key elements of the life cycle approach to regulation. Today I want to persuade you that post-market monitoring and post-marketing safety trials should be designed in such a way as to make it possible for researchers and regulators to identify potential sex-based differences. This is not routine practice at the present time. It should be.
We know that there are medically relevant biological differences between men and women, boys and girls, with respect to both the pharmacokinetics and pharmacodynamics of medications. That is to say that the same medications can and do affect women and men, boys and girls, differently.
Consider, for example, two documents that came out in October of this year — October 3 and October 9. It is the "Health Canada Endorsed Important Safety Information on ZOFRAN." ZOFRAN is a prescription medication for the prevention of nausea and vomiting for patients undergoing chemotherapy and radiation for cancer and for patients who have had surgery. The safety information released by GSK and Health Canada reports that when used in high doses, ZOFRAN can affect the "QT interval." The QT interval is the time between beats that the heart needs to recharge itself.
We know, and in fact we have known since 1990, that after puberty the QT interval is fractionally longer in women than in men, and that this difference makes women more vulnerable than men to medication that prolongs the QT interval. The "Safety Information" released by GSK and Health Canada made mention of several revised dosing instructions, including the need to "Avoid ZOFRAN if you have congenital long QT syndrome," but does not made mention of the fact that this risk is higher amongst women than men. ZOFRAN is widely prescribed in high doses by oncologists for women receiving chemotherapy for breast cancer.
Therefore, one of the first recommendations I will come back to is that all safety information letters should include a comment on sex-related differences, either confirming that there are no differences and explaining this, or alternatively telling us what those differences are. Either way, we need to know.
My second example concerns therapeutic products in pregnancy. Most therapeutic products are not tested in pregnant women and therefore are not labelled for use in pregnancy. For this reason, almost all drugs and vaccines used in pregnant women are "off-label." Off-label prescribing occurs when a physician prescribes a drug for an unapproved indication, or prescribes a drug for an approved indication but in an unapproved population, for example, pregnant women, at an unapproved dose or in an unapproved form of administration. As has already been mentioned this morning, it is legal for physicians to practice off-label prescribing. It is not legal for manufacturers to promote off-label prescribing.
Others have argued for the obligation to pay attention to this patient population. I will not rehearse those arguments here, but I have provided a reference.
Notwithstanding this moral obligation, however, drug use in pregnancy is complicated. Why? It is complicated because of the potential harm to the fetus and the newborn. Post-marketing safety for this patient population requires long-term safety studies and patient registries for both women and children. Untoward side effects of therapeutic products may not show up for years. Consider, for example, our experience with DES. Lest you think this is old history that I am rehashing, let me tell you about a current potentially risky use of progesterone supplementation in early pregnancy.
In North America, progesterone is routinely used in high doses in in vitro fertilization, usually for the first 12 weeks of pregnancy, to prevent preterm fetal loss. What evidence supports this practice? One randomized controlled trial reported in 1992 that involved 120 women. Since then, there is some evidence of harm to male children if the progesterone is not micronized. Because progesterone is a sex hormone, it stands to reason that reproductive organs could be affected. Male children have been born with penile abnormalities, but what about the possible long-term harms?
The second point is that responsible post-marketing research involving pregnant women has to include mandatory long-term follow-up, especially of children.
My last point has to do with the last population: therapeutic products use while breastfeeding. Consider here the March 2 and March 7, 2012 "Health Canada Endorsed Important Safety Information on Domperidone." This prescription medication is for the treatment of stomach and intestine problems such as gastritis. It is also used off label for breast milk supplementation to help with breast milk production in women experiencing insufficient lactation. Earlier this year, the Motherisk program at the Hospital for Sick Kids reported, on the basis of a meta-analysis of relevant trials, that domperidone increases breast milk supply, therefore confirming the off-label use.
The safety information issued by Health Canada and manufacturers of domperidone this year reports on an association between the use of the drug and the increased risk of serious abnormal heart rhythms and sudden death from cardiac arrest in patients taking more than 30 milligrams a day or in patients who are more than 60 years old. As a direct result of this warning, lactation consultants are now reluctant to recommend the use of this medication to women, despite the fact that the dosage does not exceed the 30 milligrams a day and despite the fact that the patients who are lactating are not, in fact, over 60 years of age.
This could be seen as problematic, given the benefits of breastfeeding. This is especially important for women who have had children born premature, who are not able to breastfeed immediately and have to try to maintain their milk production for several weeks until the child can breastfeed. The safety information letter should address this issue and provide direction to those who would either continue, or not, to prescribe this.
In conclusion, let me make a few very quick points. When serious safety issues are identified and there is a letter from Health Canada and the drug manufacturers, there should be explicit mention for each of these populations, either telling us that we should also be concerned about them or telling us there is nothing specific to those populations that we need worry about. If we are going to do post-marketing trials, we have to be serious about long-term follow-up, and we have to do that for both the pregnant women and their offspring. If there are to be treatment registries, there is a special challenge for regulators, then, in determining the time frame. How long will the follow-up be? In the ideal world, you would want to actually go on to middle age. In the real world, I know that is not possible, but what then will be an appropriate time frame for long-term follow-up? Also, all safety information should provide providers with the basis of that recommendation, so access to original sources.
Lastly, I want to end with a recommendation that is not my own but also made by Thomas Moore:
For the first three years after approval, new drugs should carry a special warning akin to the black triangle used in Britain. It should be prominent and mean to every physician, New Drug: Caution Indicated.
The Chair: Thank you.
Now I will turn to Mary Wiktorowicz, Chair and Associate Professor, School of Health Policy & Management, Faculty of Health, York University.
Mary Wiktorowicz, Chair and Associate Professor, School of Health Policy & Management, Faculty of Health, York University, as an individual: Thank you for inviting me here today. In my presentation I will address three key points.
First, public confidence in Health Canada has eroded due to its lack of transparency and due process for public representation in the drug review process. I believe a new strategy for pharmacovigilance is needed, and I will offer insights from our research on international governance models to guide a more systematic approach. Active surveillance should be arm's length from pharmaceutical companies, whose research results are frequently biased.
Let me begin by saying that adverse drug reactions are one of the top 10 leading causes of death. This is based on results from the U.S., the Office of Inspector General, in which observational studies have been done in hospitals.
Patients are increasingly exposed to the harms of medicines. For example, markets for blockbuster drugs are so large that even risks that occur infrequently can lead to thousands of injuries, including the deaths and cardiac events associated with Vioxx and Bextra. In fact, I would say that it is not just a crisis of confidence; it is a rolling crisis of confidence, because every year we hear of additional products that have problems and a series of lawsuits in the tens of millions of dollars and more. It seems that the pendulum has swung. It is no longer the regulator where the penalties are happening; it is through the legal system. I think that should be of concern to everyone.
Pre-market drug review is recognized as incomplete, and there are many studies that show this. It is generally recognized, first, that the brief randomized controlled trials do not detect harms from long-term use. We see this, for example, most recently in the product olanzapine, an antipsychotic, where the trials are perhaps six months, and people stay on these drugs for the rest of their lives. What they found is that people gain weight, they acquire diabetes, and now there have been huge lawsuits against the manufacturer for these reasons.
There are too few subjects in clinical trials to identify rare side effects. Generally, you need about 1,500 subjects in a trial to identify a rare side effect.
As the other speakers have noted, due to off-label prescribing, the market for drugs can include patient groups that were never tested, leaving these patients at particular risk: children, youth, seniors and women.
Demonstrating efficacy in a randomized controlled trial sufficient for market approval offers a fairly low threshold. It is basically short-term improvement in a surrogate marker compared to placebo. However, this question often fails to answer doctors' and patients' more relevant questions about the long-term effectiveness and mortality.
The other side of the coin is that once the product is out on the market, spontaneous adverse drug reporting is passive, slow, expensive and incomplete. I would like to site one example.
The FDA received about 82 adverse drug reaction reports for the drug digoxin in the 1990s and it did not seem to be a big problem. Following that, a systematic survey of U.S. Medicare records over the same time period — early to mid-1990s — found 202,000 hospitalizations for adverse drug reactions related to digoxin.
The point is that if we are waiting for these spontaneous ADRs to come in, first, less than 10 per cent come in; and second, you can actually go out and look for these when we have the electronic administrative databases that could get those answers and strengthen those signals for ADRs much more quickly.
At the same time, Health Canada's processes are not transparent. The fundamental point of a regulatory agency is transparency, to give the public, at whose behest the agency serves, confidence that it ensures due process in fulfilling its role. Regulatory agencies such as the CRTC ensure public representation within their mandate. Health Canada obscures its regulatory process from public purview and disallows a public perspective within its regulatory decisions. The Auditor General has found that Health Canada is not adequately fulfilling its responsibilities, and very recently The Toronto Star's reports attested it is losing public confidence.
In terms of the harms associated with the drugs for ADHD, for example, 9 per cent of children on drugs for ADHD attain psychotic symptoms. Neither patients nor parents are informed of this, and doctors are not informing them either.
I am fairly close to my conclusions. One point I would like to make before going to them is that Health Canada's health product vigilance framework notes that it will use risk management plans as a vigilant strategy. However, there are severe problems with risk management plans. Giezen's analysis of 18 risk management plans in the European Union identified 169 safety concerns. Frau found that risk management plans were used to reassure the public as inadequately evaluated drugs entered the market.
Essentially, risk management plans are developed by the product sponsor, the manufacturer, approved by the agency, and it is the manufacturer who runs the risk management plans. Sometimes there is no more to it than informing physicians about the product; sometimes there are observational patient registries, but those registries, from a scientific perspective, are flawed because they do not include a control group, so you are not getting scientifically rigorous information from them.
I wanted to go through some of the comparative studies that show, for example, that the FDA engages multiple organizations in its post-market surveillance process. In the package I have provided you will see a couple of diagrams of the FDA engaging with Veterans Affairs to do some data mining to determine the risks of drugs that are being used. In New Zealand, for example, they have a medication monitoring unit where industry is not doing the post-market studies; it is the regulator, Medsafe, that works with the medication monitoring unit at the University of Otago to identify the risks of products.
The FDA has a Drug Safety Board that includes the Centre for Drugs' directors as well as members from other government agencies including Veterans Administration and the Department of Defense, both of which have big health care systems; the Agency for Healthcare Research and Quality, the National Institutes of Health and perhaps some others. The board meets monthly and looks at specific safety issues and also reviews how the FDA is communicating these messages to the public.
I believe a new strategy for the governance of pharmacovigilance is required in Canada. Membership on the new pharmacovigilance committee should follow the FDA Drug Safety Board model where, in addition to Health Canada, there would be representatives from the Drug Safety and Effectiveness Network, provincial government drug plans and experts with access to health care databases that can support active surveillance.
Of course, the funding for DSEN, whose work is extremely valuable, should be increased.
I have more comments on Health Canada's processes, but because we are pressed for time I will move on to my conclusions.
I believe a new strategy to address the harms of pharmaceutical drugs in Canada is needed and that this new governance strategy should be a process where Health Canada collaborates with provincial counterparts and the Drug Safety and Effectiveness Network to access data on the effects of medicines in the real world and has the authority to address issues with medicines. The funding for DSEN to commission independent studies to support Health Canada's regulatory decisions should be increased, and Health Canada should be equipped with the authority and resources to fulfill its responsibilities. More transparency should be required, and public representation should be included within the review of pharmaceutical products, not just advisory committees to Health Canada. Finally, the safety of Canadians should be prioritized higher than the proprietary interests of industry, which currently seems to be one of the main inhibitors to greater transparency and greater sharing of information.
The Chair: Thank you all very much. I will turn to colleagues now for questions.
Senator Eggleton: Thank you all for your presentations. You have been very helpful to this committee today, as you were in the past in the first phase of our study.
I will focus first on the Drug Safety and Effectiveness Network because I think you see some value in it and believe that it has helped in terms of the services it provides, but you also have some reservations.
Professor Lemmens, what would you do to strengthen it? Would you separate it more from CIHR? You have expressed a concern about CIHR having two members from the pharmaceutical industry on their board now and talked about what that can do to the perception of independence in terms of the work of DSEN. Could you comment further on how you would change DSEN as it is now?
Mr. Lemmens: Thank you for that question. I think that there is indeed significant value in the federal government's initiative to set up the Drug Safety and Effectiveness Network and to provide it with what is, at first sight, a significant amount of funding for a significant period of time. However, as I already pointed out, it is obviously an easier recommendation to make if you are not at the government level. I would say it needs solid funding, and it certainly needs not only to promote independent research but also to be seen to be structurally independent from industry as well as from Health Canada.
There is in the literature quite some emphasis on the fact that in a regulatory framework where an agency is primarily focused on the approval of products on the basis of pre-market data there is a certain potential incentive for not wanting to have to admit mistakes that are being made in the context of the approval process. Many people have emphasized that there should be a monitoring agency independent from the agency that approves the drugs, because there is a difference between post-marketing surveillance and approval based on pre-market data.
I think that structural independence from Health Canada is needed, but also, in my view, from the CIHR in light of the increased focus of CIHR on direct collaboration with industry, particularly in the context of clinical trials. I am not against interaction. Obviously industry is an important player in the context of drug development, but I think if we look at the problems that have been identified in the conduct of clinical trials and that have come to the fore in various controversies and in lawsuits in the United States and prosecutions by the FDA, it is clear that there are some inherent pressures in the system. If CIHR focuses particularly in the context of drug development on close collaboration with industry, there is a certain overlap of interest that occurs, so I would say yes, there has to be more of a separation. Either CIHR itself has to go back to the drawing table about how it structures its relation with industry and creates more structural independence from industry or the DSEN has to be more independent from CIHR.
Senator Eggleton: I will pick up on something that you once proposed, because it fits in with this. In an article entitled Regulation of Pharmaceuticals in Canada you described a proposal of creating a new agency that would be comprised of three branches: one for drug approval, the second for post-marketing studies, and the third for drug information. Do you think there needs to be even more independence than DSEN? You were suggesting that something broader than that would perhaps be more beneficial.
Mr. Lemmens: Yes, the DSEN is an intermediary solution. I support claims that have been made by others in the literature. In the article I make reference to authors who have proposed in leading medical journals in the U.S. as well the establishment of an independent drug testing agency. It does not need to be seen as a complete state organized clinical trial business. You can think of a system whereby, instead of the pharmaceutical industry contracting out to clinical trials, the government fits in between as an intermediary and decides to empower independent clinical trial entities to conduct the research that it has approved.
Senator Eggleton: Professor Wiktorowicz, you have also talked about the public confidence in Health Canada being eroded and about biased work, vested interests, et cetera. How would you see these elements, whether it is the Drug Safety and Effectiveness Network or the whole system, becoming more independent?
Ms. Wiktorowicz: There needs to be some separation from the agency that approves the drugs and the one that looks out once products are on the market because of what Mr. Lemmens just mentioned, which is that those who approved do not want to be seen as not having done their job well. However, the fact is there is such limited information about drugs that I do not think you can necessarily always fault them. It is not a question of fault; it is more a question of separation of roles because I think it is a different role.
The clinical trials are run by industry right now. Whether that would ever change, I do not know, but the question is that in the post-market phase many of the observational studies using the public databases, the health care and the drug formulary databases can be done by independent investigators. I see it as a progression of steps. For example, you get a signal from ADRs, for example, the spontaneous adverse drug reaction reporting. The regulator does not know if this is in fact something of significance or just noise. The next step would be to do some mining of databases. The steps I am going through are in fact what were described to us by a senior director at the FDA.
The next step is you would take it to the independent agency to do some mining of drug databases and see if that signal could be strengthened. If it indeed is stronger, then you might go to the next step of clinical trials.
It is a phased process, and I think it is important to have independence at some point following the initial approval.
Senator Eggleton: Professor Baylis, on this question of a black triangle as is used in Britain — we also heard that black box labelling is done out of the United States — I take it you feel that we should adopt this kind of approach. I do not understand what the difference is between the American system and our system and how you would see either or both of those applying to Canada. Could you go through that? Also, do you think it should be mandatory? Should labelling be mandatory by the regulator, who at the moment is Health Canada?
Dr. Baylis: I would like to answer your questions and then also add some comments to the others.
In terms of what has already been said my colleagues, I would recoup, re-emphasize and do it under three headings. The most important thing is independence and that is why you are hearing the comments about CIHR and about industry. If I can go further, I also think it is about Health Canada. I think you really need complete independence and you also need survivability.
I say that to you as someone who has had three Governor-in-Council appointments, I was on the Canadian Biotechnology Advisory Committee, I was on Assisted Human Reproduction Canada and I was on CIHR. The only one that has survived is CIHR, and I do not think that has anything to do with me one way or the other, but the point I am making is that it has to be a structure that is not subject to the whims of government regardless of who is in power. It almost has to be insulated the way the Auditor General is. You cannot just get rid of them because you do not like whatever it is they are coming up with.
The second point I would make is that it has to have a proactive mandate and not a passive mandate. It is not just about waiting to see what happens in the field and getting your adverse events reporting. Part of what should be happening in a post-marketing context is well-designed, randomized control trials looking for information that you did not have in the context of going forward. That is why you have heard us all refer at different times and in different ways to the off-label use. Some of that use can be very good but it gets to a point where you have to do a randomized control trial. The example I gave you of domperidone fits that nicely. It is not that off-label use is necessarily a bad thing, but at some point you do actually have to go back and test it out, so I would like it to have a proactive mandate.
The third thing is it has to be properly resourced. Whatever you do, do not forget my plea for long-term follow-up, and meaningful long-term follow-up is not 18 months. We understand that in the context of the trials that happen to get something to market, it is going to be short, but once you move into long-term you have to have a plan. It cannot just be, "Well, we will follow it for 18 months because that is all the money we have." You have to understand, for this particular drug, where you think it might reasonably have an impact, and that will tell you something about the time frame you need to follow the drug.
To get back to the issue of the warning, the black triangle is different from the black box warning that you hear about. The black box warning is the equivalent of what Health Canada does with its safety information letters. Canadian physicians will refer to this as a black box warning. It is in fact literally a black box that shows up on the labelling that everyone can read that says additional information either about dosage or time frame or new information as they have found it. In the U.S. that is referred to as a black box warning and we have Health Canada safety information letters.
The idea of the black triangle is almost using some of the symbols we have in the context of good housekeeping and doing your laundry. You look at the label and there are symbols that over time mean you do or do not put it in the wash, you do or do not add bleach or whatever, and here it is trying to have a really clear warning just the way you might have the skull and bones, or hand, or whatever. I am trying to use both negative and positive images.
The idea is it is to be helpful. It is a quick way of giving information. That is also why I said that for any of the warnings that come out — in addition to requiring this kind of marketing label so people can understand, even consumers can see that it is a new drug — you need have a way for people to go back and get access to the original trial data. If I go back to the example of the Domperidone, a person who was in obstetrics and gynecology who understood the use of that drug off-label in the context of lactation might go back and say, "I understand the data and look at it myself, and we do not have to worry about our continued use off label." It is important to go back and mandatory, absolutely. It has to be mandatory.
Senator Eaton: I believe what you are all saying is that pre-market and post-market study should be equally emphasized so that you have pre-market before the drug gets out there, and then what we should have is equal emphasis on post-market study. Am I saying that correctly?
Ms. Wiktorowicz: I agree with that.
Dr. Baylis: To tweak it a bit, I would say it is not just equally emphasized but equally robust. The reason for that slight word change is because all the examples I and my colleagues have given have focused on the safety issues. We have not focused very much on the effectiveness in the real world, which is what post-marketing is supposed to be about, because there is no incentive for anyone to go and look for what kind of real difference it is making.
I would say as an aside, in the context of a government-funded health care system, there should be a huge interest in doing that part but we tend to focus on safety. I think it is important that it go beyond safety and adverse events.
Senator Eaton: I understand your comment.
Professor Lemmens, you said that Health Canada has been working since 2005 on a progressive licensing framework that, among other things, would enhance regulatory oversight of the post-market phase and give the government authority to require drug manufacturers to conduct post-market studies and to submit the resulting data for review. Health Canada has obviously not done that yet, have they?
Mr. Lemmens: It has not finalized it and there are obviously political reasons for not finalizing a project. We have had elections in between, but I would say it should be hopefully a priority of Health Canada. Health Canada has certainly indicated it is continuing to work on this, but this has been ongoing for a long time. If we look at the U.S., we see that in the wake of serious controversies that highlighted the significant risks of the fact that the approval system has limits and does not necessarily detect small problems of drugs that can have massive consequences once they are prescribed, the U.S. has taken action. Since 2007 it has introduced much more stringent post-marketing requirements and has given many more powers to the Food and Drug Administration to control and to request post-market surveillance studies from manufacturers. I would say Health Canada is working on it but I certainly hope the government makes this a priority and introduces stringent powers that Health Canada can exercise to request additional information once drugs are on the market.
I will also express a caveat. There is one concern that people have. If you look at the history of drug regulation, there always has been a pendulum swinging towards more regulation and less regulation. There has always been a tendency to bend interesting initiatives to some degree as a result of industry pressure.
I think we should not reproach this to industry. Industry has a certain interest, for example in getting drugs quickly on to the market. The concern about the new proposals to have a better, progressive licensing system is that the emphasis on the approval stage will diminish. I would say that it should not lead to a diminishment of the approval stage or to a weakening of the required evidence for the approval of new drugs, but it should be a system whereby additional powers are given to Health Canada.
Senator Eaton: They should be equally weighted, in other words?
Mr. Lemmens: They should be equally weighted and they should not only focus, as others have pointed out, on clinical trials but also on how drugs are used in the real world and how adverse events can be better detected.
Senator Eaton: Professor Baylis, when we talk about the real world use of drugs, your emphasis on women is interesting to me because we all saw those ads a couple of years ago saying that women have different symptoms for heart attacks and strokes and that if you are a woman, these are the symptoms you must look for, which are different from a man. I understand why we would react differently to drugs.
This is a very interesting direction. Do pharmaceutical companies do clinical trials to test women to see if there is a different reaction than with men and to ensure that that reaction is noted, or is it just generally taken, from a population basis, on an average?
Dr. Baylis: One thing that is really important to know in the context of the history of clinical trials is that it is really only in the 1990s that there was a strong push to include women in research. In fact, the literature has somewhat sarcastic titles to the effect of "only white, single men between the ages of 35 and 55 need apply." In the 1990s, you had a huge push to say that women's bodies are different. We metabolize drugs differently, we go through puberty and we have different ways in which our body responds to different chemicals, et cetera.
You have now seen a correction, to be fair — at least with respect to women. They are included in greater numbers, but we have two other new problems that have surfaced. One is that they are included, but their data is not analyzed separately. You have met the pro forma requirement of "including women," and in some trials they will be 50 per cent of the population, but you do not then look to see if there are important differences.
Senator Eaton: The data is averaged out, in other words?
Dr. Baylis: Yes, the data is all presented together as though they were all the same bodies. That is one problem.
The second problem, which I will not elaborate on today but would put down for notice, is that women are required to take mandatory contraception. I have a number of grave concerns about that.
The Chair: We dealt with those things in detail in our previous study. I have allowed this, just to inform the senator with regard to her questions, regarding post-approval monitoring and not on the clinical trial part.
Dr. Baylis: Specifically it is building on Mr. Lemmens' response to you. I am aware that Health Canada has had some invitational workshops specifically on the plan that it has been developing for last seven years. The last documents I have date from 2010. I am happy to share those with you. My copies are all marked up, but they have a topic on market authorizations, which is "Issuance, Contents, and Standards and Special Conditions." I would draw that one especially to your attention in light of my concern, which is that in the context of their market authorization, they have a category of standard conditions and a category of special conditions. That is, some drugs would be approved for post marketing with standard conditions. I want to signal that "special conditions" are things like requiring confirmatory studies of a product's benefit and long-term safety studies. I do not understand why those are special conditions. I think they ought to be standard conditions. I would encourage you to look at the work they have done with a critical eye.
Senator Eaton: On adverse drugs reactions, do you see a greater role for pharmacists in all of this in trying to cull when they give out a drug? Do you think we educate pharmacists to be more proactive in educating people about what they are putting into their systems, what the reactions could be and how, if you have this reaction, you should come and tell me about it and about what other drugs are you taking? Are we doing enough on that front? Are pharmacists playing a big enough role?
Dr. Baylis: Last weekend I was at a meeting of care on recommended but unfunded vaccines. In that context there was much evidence, including anecdotal evidence, presented to the effect that pharmacists do a wonderful job, a much better job than clinicians, partly, it was said, because they had more time. The rule we use is six minutes. You have six minutes to move a patient in and out. That is not a lot of time to have a conversation about anything.
With respect to vaccines, a number of provinces have tried to move that into the realm of the pharmacy because they are knowledgeable and capable. There is something there worth exploring, I think.
Ms. Wiktorowicz: I want to emphasize Mr. Lemmens' points on progressive licences, and Dr. Baylis also mentioned it. The idea of progressive licensing is that some products will appear to be not as harmful so you would require not three clinical trials, perhaps only two, and then you might monitor them in some way. I think that lowering of pre-market safety standards would diminish public confidence in Health Canada.
I have concerns around progressive licensing if that is the path that Health Canada is going to take. That is the path that they have documented that they would take because they see it as a life cycle approach. However, we are already not getting enough information in the pre-market stage, so to reduce that amount of information is a concern.
Senator Dyck: I would like to thank all our presenters this morning for their concise, precise and very helpful information.
The discussions this morning have been very helpful. We were talking about what happens in the real world versus the ideal world. In the ideal world, scientific objectivity is our goal and we achieve it, and patient safety takes priority over the interests of drug companies. However, we know that we live in the real world and that pharmaceutical companies have interests that may, in some way, limit our scientific objectivity in some circumstances.
The question that intrigued me with regard to adverse drug reporting and post-market surveillance was with respect to drugs that are on the market and that may not actually do what they are intended to do. They actually may create harm because they are not active. For example, I was thinking of psychoactive drugs. Ms. Wiktorowicz mentioned olanzapine. However, what if there is a psychoactive drug that is an antidepressant, but it is ineffective and the adverse drug reaction could be worsening of depression or even suicide? How do we follow up non-effective drugs? Do we follow them up? If so, how do we follow them up? How would a system pick up a drug that is not being effective? Would that be recorded within our monitoring?
Ms. Wiktorowicz: Yes, I think it is a problem. Drugs are usually approved on the basis of something referred to as a surrogate marker. If I could use the example of statins, which are cholesterol-reducing drugs, they are approved based on their ability to lower cholesterol in the blood, but the final effectiveness is assumed to be lower cardiac events, lower strokes, and all those kinds of things. However, there have not been clinical trials to show that at least at the pre-approval stage, that is, in fact, the case.
We have these surrogate markers, and lately we have been having more and more trials around the lipids. We have been finding that they seem to show effectiveness for secondary prevention, which is that once someone has had a heart attack, and then if you are put on that drug, they may prevent a second one. However, often physicians prescribe statins for people who simply come in, their blood is tested and they have high cholesterol. For primary prevention, statins are not effective. There is great controversy out there because they do have side effects in terms of pain in the muscles. Not only should we be exercising, but if you have pain in the muscles, you will not exercise, so the person on these statins is worse off.
That is a long way to say that often the long-term effectiveness and the actual real world effectiveness is not properly tested unless there are investigators out there willing to do those studies. The NIH or the CIHR would fund those.
Senator Dyck: I want to follow up on the idea of different drug effects with different populations. We were talking about gender, the differences between men and women. We all know there are marked differences between the different races in terms of drug-metabolizing enzymes. For example, if a drug is one that is unacetylated, are there any warnings on certain prescriptions that if you belong to such-and-such a group that is a slow acetylator, this may not be the drug for you? Is there anything like that in the real world, or should something like that be proposed as directed toward target populations?
Dr. Baylis: There certainly have been. Following on the sort of admonitions to include women in clinical trials, there were similar concerns about diversity and different racial and ethnic groups not getting appropriate representation. The problem remains that even when the issue of representation gets addressed, it is the issue of analysis of data by those different streams that is still not happening in any kind of regular way. Typically what happens is that we end up learning this by trial and error, over time, in the real world, after the fact. Yes, there is an ongoing issue or concern there.
The Chair: Mr. Lemmens, do you have a comment on either of these two questions?
Mr. Lemmens: It may be to simply highlight again the importance of having a regulatory agency look at what is happening in the real world of drug prescription. I can mention here a 2012 study. I have a reference in the memo that we submitted. This study revealed that about 11 per cent of prescription drugs in Canada are prescribed for off-label use, and of these off-label prescriptions, 79 per cent lacked strong scientific evidence of the off-label use. That means that one in ten drugs is prescribed off label, and 79 per cent of these prescriptions lack a solid scientific basis.
A study in the United States in 2006 found that 21 per cent of prescription drugs were prescribed off label, and of that number, 73 per cent had uncertain or inadequate evidence that that prescription would work.
You mentioned certain types of drugs. Off-label use, in the Canadian study, was highest for central nervous system drugs. Twenty-six percent was prescribed off label — anticonvulsants, 66 per cent off-label prescription; antipsychotics, 43.8 per cent; antidepressants, 33.4 per cent — all prescriptions for which there is no solid scientific evidence available.
This actually shows that we are relying on a regulatory system that focuses at this point on a one-time assessment, on the basis of very limited clinical trials evidence, of safety and efficacy, and after that it is the free world. We actually should emphasize that some regulatory agency has to look at what is happening in the context of current prescription rates, and certainly the medical professions have an important obligation there as well. It is certainly not only Health Canada. People think Health Canada does more than it does. It approves a drug, but after that drugs are prescribed in various ways that are highly problematic and that can be very harmful to Canadians.
Dr. Baylis: I have a quick comment in response to Senator Dyck's first question. If I go back and look at what Health Canada is proposing with respect to this program of market authorization, they say — and I believe they still believe it — that once they receive and review an application for market authorization, they would grant this "if Health Canada is of the opinion that the person" — and that could be the manufacturer — "has established that the benefits of this drug that are associated with the therapeutic product outweigh the risks."
My point in response to that is that what has happened is that we have all diverted our attention to the risks, with the idea that if we get evidence that the risk is going up, then we think we have an imbalance in the harm/benefit ratio. However, logically you will get that same imbalance if the harms are constant. Therefore, they have made no mistake about the harms, but the claims about the benefits are actually not as robust. Over time, we realized there are no benefits.
That is one of the problems with the way the current system is set up. It gears all of our attention to only look at the safety part of the equation, and the real world is supposed to be looking at both. I know that is a huge problem.
Senator Dyck: This just occurred to me when Mr. Lemmens was speaking. I do not know whether anyone has considered or does post-market approval of withdrawal from CNS drugs. We know that the brain is plastic and there are receptor regulations going on and different things happening. When you go off CNS drugs, some of which are meant to be temporary, do we monitor the adverse side effects that happen when you go off them?
Mr. Lemmens: I think it depends on the surveillance that is exercised by the physician prescribing the medication and on his or her reporting habits. I am not saying there is no monitoring, but there is certainly insufficient monitoring in many of these cases as to what exactly is happening.
Senator Seidman: Professor Baylis, the last time you appeared before the committee you discussed the importance of including pregnant women in clinical trials. Today in your presentation you said you really want to persuade us that post-market monitoring and post-market safety trials should be designed in such a way as to make it possible for researchers and regulators to identify potential gender-related differences, which is not routine practice.
All of you have referred to the FDA and their most recent changes and new powers. Indeed, they were here a week or two ago and discussed those with us. I would like to know if any of you could give us guidance on this and if you think that this might be a model for Health Canada in order to give them the ability to evaluate and ask for post-market studies on new drug approvals or to force label changes or things of that sort.
That is obviously a question that talks about a model for Health Canada, but I would also like to, in the second part of the question, pick up on what Professor Baylis said, perhaps twice now, namely, that we are focusing on the issue of safety, not on the issue of effectiveness. Perhaps you could elaborate on that a bit. As you say, that is an important aspect. You even make reference in your presentation to the absence of incentives for manufacturers to look for and report on additional data about effectiveness.
Dr. Baylis: I will take the second question and leave the other one to my colleagues.
Part of what I am dealing with is the fact that money is limited, and therefore people worry about harms, and harms are the things that take you to courts and adjudication, and they make the press.
Lack of effectiveness basically means we are wasting money, but it is two different bodies. The people who are wasting the money through the lack of effectiveness are us, all Canadians, taxpayers, who are contributing to the health care system. If you are looking at where is the incentive to go looking for the efficacy data, it really rests with government in some sense. That is also one of the things that underlies the concern about trying to ensure that you have an organization with independence but also a broad enough motivation to look out for both the safety and efficacy of a drug intervention in the general population.
The only other thing I would say is that the concern I have is shared broadly in a number of regulatory frameworks, because there is no incentive for someone who is making a profit off a particular product to say, "With five years' experience, we realize that we are actually 1 per cent better than the best thing out there on the market." They might in fact be very effective, but if you were the person paying for that drug, as the government, you might say, in terms of an effective use of public dollars: This new drug, which is 1 per cent better, is actually 50 per cent more expensive. We will take the consequences of losing 1 per cent of efficacy in order to save 50 per cent, which we will redistribute in the health care system.
That kind of decision making happens routinely in New Zealand where they have bulk decision making for the country as a whole. There again you have an incentive to truly understand the efficacy part of the component because you might actually make some trade-offs in a government-funded health care system.
Ms. Wiktorowicz: I will address whether Health Canada should have the power to require post-market studies. I think, most definitely, yes. However, it is not just the power to require them; it also has to have the authority to fine and issue penalties.
What has been found in the U.S. and Europe is that they have those powers, but when times lines were set for post-market studies to be completed, half of the studies never were. They might have started, but the industry was not meeting the time lines. That would mean authority to request post-market studies and issue fines.
I would like to add something else. When manufacturers submit clinical trial data, in Canada often they only submit summary data. What has been found in the U.S. and the U.K. is that those regulatory authorities have been requiring submission of raw clinical trial data and they reanalyze the data.
For example, the adverse drug reaction of suicidal ideation in SSRIs was identified that way. The manufacturer did not identify it; it was the U.K. expert Committee on Safety of Medicines that identified that adverse drug reaction, and then the FDA pooled all of their randomized control trial data and found it. Within individual clinical trials the numbers were not high enough to pick up that ADR.
Right now Canada has the power to request the raw data, but often I understand it does not and just accepts the summary data. That data really should be reanalyzed. There is the expression, "lies, damned lies and statistics." Industry can look at that data from its own perspective.
Turner et. al. in the U.S. that did reanalyze U.S. clinical trial data submitted to the FDA found that manufacturers overstated the effectiveness of their products and understated the harms. You are not getting a true picture when you just accept what the manufacturer is submitting.
Mr. Lemmens: I have a brief comment to emphasize, something my colleague Françoise Baylis pointed out and I also indicated in my presentation, which is the importance of comparative trials. There is obviously a provincial-federal jurisdictional issue here. The provinces are in charge of determining drug funding and they have the primary interest in determining what drugs are more effective than others, but there clearly should be better collaboration. There is already collaboration between the provinces, but there should be better collaboration between the federal and provincial governments and, to the extent that it is possible because of this jurisdictional division, better coordination of the control of safety and basic efficacy, but also comparative efficacy of trials.
My colleagues Colleen Flood and Patrick Dyke recently published a paper, which I would recommend to the committee, where they discussed this. The paper is entitled The Data Divide: Managing the Misalignment in Canada's Evidentiary Requirements for Drug Regulation and Funding and focuses precisely on this problem and highlights the importance of obtaining comparative data.
I would also highlight that this is not only a financial issue, but also a safety issue. Currently we can have drugs that are inferior on the Canadian market being prescribed for a particular treatment, whereas another comparative superior drug is also available on the market. There is a patient safety issue as well at stake in the lack of comparative evidence.
Senator Cordy: I would like to talk about the lack of transparency in the post-approval monitoring system. Should we know or do we know as patients, people taking the drugs, what demographics have been part of the initial clinical trials and post-market studies? Should we know or do we know if pharmaceutical companies are part of post-approval monitoring? We heard from the Professor Lemmens earlier that CIHR has just appointed two pharmaceutical executives to the board.
How can Canadians be reassured that the post-approval monitoring is independent and that it is fair and we are getting all the information we should be getting? I believe the FDR has a number of meetings open to the public, but in Canada we do not have that luxury.
Mr. Lemmens: I will say briefly that the transparency issue was already discussed in the context of clinical trials. Certainly, in the context of post-marketing surveillance, greater transparency is clearly required. There are some initiatives that have been taken recently by the Minister of Health with respect to the development of a registry for clinical trials.
Unfortunately, in my view, the registry requirement is insufficient because it does not include a strong regulatory requirement to register and publish the results — not only to register the clinical trials but also publish the results of clinical trials — which should be in the pre-marketing and the post-marketing stage. Currently in Canada we do not have a strict regulatory requirement to register and publish the results of clinical trials, which I think is clearly lacking.
If we want to go to a more transparent and trustworthy post-marketing surveillance system, that would be an important step. It would not solve all of the problems, but certainly a registry system with mandatory results reporting and sanctions associated with it would be an important first step that Canada has not taken. This is a step that has been taken in the United States, in Argentina and is currently also being debated in Europe.
Dr. Baylis: I would answer your first question about ensuring independence in this context by simply saying that you have to make it independent. What you have right now, and you have heard all of us say it, are structural conflicts of interest. From my perspective, that is not calling into question or impugning any individual. That is about saying structurally you have set up institutions that have a clear mandate and then you give them other jobs or they take on other jobs which are not theirs.
What is the core business of CIHR? It is to fund research, not to be a regulator. That job, looking after the interests of Canadians, belongs somewhere else. That would be my first comment. If you want it to be independent, make sure it is structurally independent.
In relation to your original question about patients, what they know and sources of information, how many of you had seen the Health Canada safety information letter before I brought it to you? If you look at my presentation, in every case I gave you two dates. Why? The first date is the date for the document that goes to the health care provider. The second date is the date that goes to the patient, consumer, presumably. I do not know of a single patient who would know to go looking through the Health Canada website to find this letter. I do not know of a single physician who, when they receive their letter, anxiously awaits the patient version of the letter so they can distribute it to their patients. I am not saying they do not, I just do not know that to be common practice.
We need to remember that patient sources of information are, by and large, their physicians, and we need to empower physicians. Physicians are in the context right now of very stressed health care delivery. One of the things we know is very often, when patients get on medications, they get more and more added to them and it almost takes a change in physician for someone to sit back and reassess what they are taking.
There are huge challenges with respect to how to make information available to patients.
Ms. Wiktorowicz: I will respond to your question around transparency. The Drug Safety and Effectiveness Network, which is doing some of these post-market studies now, is transparent. They have everything on their website, pretty well. They have a great network of nodes of research, with specialization and so on. That is very transparent.
I think the problem is that Health Canada is not transparent. They have a summary basis of decision and there is very little information there for clinicians, even provincial drug benefit plans, to really understand a new product when it comes out.
Second, Health Canada has indicated it will pursue the strategy of risk management plans. It is a harmonized strategy, based on the International Conference of Harmonization. As I mentioned, it is a flawed process. Industry processes how they will keep track of a drug in the post-market phase, but it is industry that develops the plan and it is negotiated with Health Canada, but there is no sense of this being a rigorous approach.
There have been critiques. We interviewed people in France, for example, where these have been going on for a long time, and they think it is an opportunity for companies to see the drug, introduce it to physicians and give experience to physicians. The studies are weak and not scientifically rigorous.
I think there is not a lot of transparency around those risk management plans, so we really need an independent body that can ensure transparency.
Senator Cordy: Professor Lemmens, you said the subtle promotion of inferior drugs can be dangerous. Can you expand on that comment?
Mr. Lemmens: This would be a topic of another hearing, but a lot of prescription in practice is determined by the medical literature. There are various reports of how the medical literature has been skewed by financial interests. Medical articles describing the benefits of drug products have often been misleading. They are actually criminal prosecutions that have been associated with that in the context of the United States where settlements were obtained, so we cannot say for sure what exactly happened.
If you have a drug on the market and you are able to steer prescription patterns by launching publications in the medical literature that influence prescription behaviour, you can have physicians honestly thinking they are prescribing the best drug when they are actually prescribing an interior drug. Clearly, if you were a patient, you would not want to be treated with a drug that is less effective than its competitor.
Senator Cordy: I wonder how we can make the reporting of adverse drug reactions easier. We heard yesterday from witnesses who talked about electronic health records being a good way to do it. Senator Eaton talked today about using pharmacists more effectively because they might have a little more time and experience with the drug reactions, and they know all of the medications a patient could be on.
Professor Lemmens you talked about doctors, but indicated, rightfully so, that it can be pretty time consuming.
In this day and age of technology, surely there is an easier way to ensure that patients can record adverse drug reactions or, as we mentioned earlier, drugs that are just not effective.
Dr. Baylis: I wholeheartedly agree. We rely on patients and their memory. "Have you ever had a reaction to. . . ?" Then you sit there and say, "I do not know; I do not remember" unless it was dramatic. Electronic health records are an important tool and a really huge challenge for Canada because of the Constitution, the way responsibility lies and because of the Privacy Act. Over time, technology has developed in ways such that you cannot necessarily even merge platforms, even there was the will.
It is a huge problem.
I do know that CIHR is investing money in a pilot project to look at how that might happen in the Maritimes to see if there is a will between those provinces to be a pilot study to figure out if we can take the electronic records, work within the law with respect to the privacy issue and be able to do so that, as Canadians travel across the country, they would be able to look it up if they have a health problem, wherever they are.
That would be excellent not only for drug reactions but for all kinds of things in pursuit of health and well-being.
Senator McInnis: As a substitute on the committee, I am loath to speak directly to the specific subjects that are being discussed here this morning. I must say, Mr. Chair, having been here yesterday afternoon and this morning, I found this most interesting.
I wanted to ask a general question because I see there is considerable mention of Health Canada in that government should play a more prominent role with respect to drug clinical trials. The credibility of Health Canada has been mentioned here, as well as a lack of transparency. The Auditor General concurs with this. Health Canada should work more closely with the provinces. Of course, I heard also "the safety of Canadians," which goes without saying should be paramount.
I have two points: What efforts if any have been made to undertake some contact or a body to ensure that there is ongoing dialogue or opened communications? What vehicle is there? Has anyone talked about this? What vehicle would bring together the federal government, the provinces, the pharmaceutical industry and the public?
I have not read the terms of reference of the committee, but has anyone given that any thought? It strikes me that this will be an ongoing problem. There is a multiplicity of issues and entities involved here. It seems to me there should be some thought as to a vehicle as to how we can resolve some of these difficulties. With respect, the criticism may well be well-founded. I wonder if that will be part of the deliberations.
The Chair: I will try to focus this question on our current study here. We are not dealing with an investigation of the interaction of the federal-provincial and all the other bodies. A number of you have spoken today about several aspects of this, and I think one of the real issues comes down to the authority, the transparency and the independence of the bodies that look into these issues.
I wonder if I could take the question and come back specifically to the testimony we have heard here today. Starting with Professor Lemmens, perhaps we can come back to the role of the DSEN, which you indicated you felt plays an important role but in fact should have perhaps a bigger role to play and so on. The real issue that all three of you have touched on with regard to the DSEN is its ability to collect and identify information. However, the question really then turns around to: Where does that information go and how do the observations that it finds get implemented?
I will make a specific question to you in that regard.
I think it might be difficult to find a way to do this, but do you suggest that the DSEN should have any regulatory authority in addition to the role that it plays now?
Mr. Lemmens: In its current construction, it clearly is not a regulatory body, so I would be reluctant to say it should become one. There should be serious thought given to how we can construct an independent agency that can oversee more independent, post-marketing surveillance. The DSEN can continue to play an important role in the context of promoting that type of research and coordinating that type of research. However, in light of the problems that we have seen with respect to the reliability and integrity of a lot of clinical trials of drugs, strengthening a system's independence, even if just for an agency focusing on providing the necessary evidence under which other regulatory agencies could base decisions, would be an important component. That could be emphasized, as well.
I am sure Ms. Wiktorowicz can say in more detail how the DSEN currently functions and how it could better promote these types of things.
The Chair: Do not go into all that it does now, but just get to that other aspect of whether it should have a regulatory role, if you could.
Ms. Wiktorowicz: I think it should definitely contribute to our regulatory role. Currently, the DSEN informs Health Canada of significant findings, but Health Canada has the right to do what it wills with that information. I think DSEN should collaborate with Health Canada as opposed to simply being a body that provides input.
I think there is a structural problem right now. For example, the U.S. has had legislation, the FDA Amendments Act, that requires the FDA to include other public agencies, such as the NIH and so on that I mentioned earlier, to have some input into regulatory approaches and decisions. Unless we require that of Health Canada, it will not happen.
However, I think it is essential because once drugs are out on the market, all the other public agencies have to deal with it. Yet, in a way, they are blind in that they do not see all the information. The proprietary interests of industry in not sharing clinical trial data are what seem to motivate Canada in not sharing information and keeping everything very closed and not transparent.
It is almost as though it has to be required to take a different tact or process.
Dr. Baylis: In the context of the original question and your amplification, it is important to acknowledge that Health Canada has tried to host some of these invitational workshops, especially in the context of some of the proposed changes. However, one thing that is important, just as my colleague has indicated, is that it is not always clear to me how effective those meetings are, because of the lack of trust and credibility. The participants come "parti pris," assuming it will play out in a particular kind of way.
It is true that there is this perception of a lack of independence on the part of Health Canada because even when Health Canada does say it will do something, it seems to bend a little with any push-back. One example was that Health Canada had proposed that if it either issued or refused to issue a market authorization, it would make this information publicly available; it would either say, "Yes, we did approve this" or "We did not," then you have biotech saying back to Health Canada, "Well, I do not see why you would publish if it did not get through." Then we would just remove that from the database of information that is available to you. Then, all that Canadians presumably would ever know is that these ones made it through the process and we just do not know but assume some did not.
Part of building up the robustness of an organization is for it to be transparent such that people could say, "This percentage of applications for market authorization was approved and this percentage was not. These are the drugs they were for." You do not have to share industry secrets in order to make some basic kind of information available.
There are different interests at play. For me, that is coming back to the main point of what you are looking for with respect to independence. You want an organization that has as its primary — if not its only — goal to worry about the health of Canadians and to ensure that products made available to them come with a very clear understanding of both the benefits and the risks because even when things get to market, they come with side-effects. We are not asking for the perfect drug that cures and has no side-effects, but we do want full disclosure and we want it in a context where people have enough information to weigh that up.
The Chair: I directed your question, senator, but do you have another question?
Senator McInnis: No, that is fine.
Senator Eggleton: I wanted to get back to something. Senator Cordy asked a question about adverse drug reaction reports, or ADR reports, which have been described by you and others as passive, minimal and inadequate.
Yesterday, Canada Health Infoway described a system they found out about in Boston. It is being used in the U.S. It was a follow-up system to check with people. It was a more active system as opposed to a passive one. I think Senator Cordy also mentioned that example.
I would like to get your thoughts on how we can improve this system. The percentages we hear are very low in terms of what we get back from physicians, the public or anyone. How should Health Canada be dealing with these? I take it that Health Canada does not. It waits until it accumulates some before it might send out a signal somewhere. Where does it send the signal; how does it communicate back to the people who filed the report and the public in general?
Do you have any thoughts on how we can — at least we are while operating under the system we are operating under — improve the effectiveness of ADR reporting?
The Chair: I will focus the question in two parts. The first part is the collection of the data and then the second part is how you use that information to be informed with regard to the specific situations.
Since I was intending to ask it in any event, I would like to pick up on the introduction of the question. One thing that we have been very clearly hearing throughout this process is that it is very difficult to find a really effective way to get access to apparent adverse reactions. How do you get the individual, the physician — whatever — to recognize the potential adverse reaction and then get that information to someone who will collect it and perhaps follow up, as the senator is suggesting? We know from the records — and you have referred to — that it is suspected that a very small percentage of adverse reactions are actually totally collected.
Yesterday we heard what I thought was the most innovative mechanism for potentially getting at this sort of thing with a pilot study that is being carried out in the United States. All the individuals involved in this study have agreed to participate. It is a clearly pilot study whereby the individuals who receive a given prescription will receive a follow-up in one month. The follow-up will be a question — an evaluation — of what has been their experience with the prescription since they received it and began to take it over the last 30 days.
It struck me at the time that this might well be a very user-friendly way of moving into the larger issue of collecting data on adverse reactions and, as Professor Baylis has emphasized this morning, the need to do so over a longer period of time.
I used the analogy that, if you take your car to the dealer, you often get a quick follow-up from the company asking you what your experience was in taking your car to the dealer. People are kind of used to responding to that kind of customer service.
I am borrowing on your efforts here, senator. I would like you to try to answer that question by referring to an example. Start with this example and then give your own input as to perhaps better examples of how we can be more effective in collecting information on potential adverse drug reactions.
Would the idea of having a kind of automatic follow-up that initiates after a person is prescribed a drug and over consequential periods of time, and for those drugs that may be taken over a longer period of time, a longer period of follow-up? Would some sort of automated process like that lead to some additional benefit in this area?
Ms. Wiktorowicz: First, internationally, 10 per cent is the highest any country gets with adverse drug reaction reporting. New Zealand gets the highest in that they are closest to 10 per cent. They do the back and forth: If a physician reports an ADR, they get in touch with the patient, et cetera. After the report has been submitted, they will get feedback that others have submitted on the same one. Those who submit feel that it is a meaningful contribution and use of their time, but still only 10 per cent.
In the U.K., they have something like the example you mentioned, and that is prescription event monitoring. Physicians follow up with a yellow card every six months, I believe, to talk about the patients in their practice that are on a certain medication and how it has been for them.
The problem they found is that once you are sent three and four of these yellow cards for different medications for different groups of patient, you stop responding to them. It sounds like a great idea, but over time it is has not been as useful as they thought.
The second weakness of that approach is something I thought of after talking with Robyn Tamblyn. I do not know if you had her here but she is an epidemiologist looking at pharmaceuticals. Her point was that there is no control group in that. You are following patients on a drug, but how do you know they really differ from people who are not taking that drug? Therefore, you almost need to not just follow those people on those drugs, but you need what they call in the U.K. a disease registry, which has patients with a particular condition — diabetes, schizophrenia, whatever — and has a subgroup on a particular med and other subgroups that either are not on a med or are on a different med. In that way, you can compare people with the same conditions.
Models like the one you suggested are out there, but they have not proven to be that successful.
Dr. Baylis: I would add a few comments to that. First, a very high percentage of prescriptions are never filled for all kind of reasons. Sometimes they are filled and they are never used — for all kinds of reasons. That is a really important fact to be aware of. We teach that to our medical students because they write a prescription and just assume they have an obedient patient who buys the drug and takes it as prescribed, but that is not what happens in the real world.
Second, you would have to be mindful of a placebo effect in that context: "I want to know it is working; I do not have an interest in not seeing that." I do not know how that model would deal with confounding factors, which are not just other medications, but other activities that would play out in terms of the drug use and other facts of life.
To be frank, I would be worried about how the system, if that were "the system," would be gamed. For example, let us look at what happened with some of the early drugs around Alzheimer's and dementia. When you were not able to show it was having a dramatic effect in terms of dealing with underlying disease or illness, the value of the drug was that it made people reintegrate into the family. If you look at the research, all the advertising became about grandma being able to sit among the family and celebrate someone's birthday and look happy as compared to how she might otherwise be. Now the success of a drug comes to mean certain kinds of social behaviours and interactions rather than what the drug might originally have been developed for.
Therefore, I do not know the details of the model, but I would say immediately those are concerns I would have.
Mr. Lemmens: Patients play an important role, and I think we should not only rely on physician reporting, which is clearly inadequate and in the current system not sufficiently encouraged. Patients play an important role. However, as my colleagues have pointed out, there are some limitations to that. New computer technology and new ways of gathering and mining the data that are being submitted can be useful and can make it more interesting to promote reporting by individual patients of side effects but there are some serious limitations as to how these data can be assessed.
Regardless, it certainly should also be true that patients have to become aware that drugs are toxic, and they should be open to identifying particular problems they experience when taking the drug.
Senator Martin: I want to add to this discussion in that —
The Chair: Please pose a question and not an observation.
Senator Martin: It is relevant to what we are talking about. May I continue?
The Chair: If it is a question.
Senator Martin: Here is the question. Would a multidisciplinary approach for this be one of the key ways to ensure that all of the health practitioners who are working in the life of the patient have the opportunity to interface? The drug can be prescribed by the physician, but it could be the counsellor or social worker who is monitoring the progress of the patient.
In your experiences, would you say that multidisciplinary approaches are key or one way to ensure that information can be gathered from all of the different health practitioners involved in the patient's care?
Mr. Lemmens: I think Professor Wiktorowicz is probably more aware of the concrete ways in which the DSEN and other monitoring programs work. However, I would say that, in general, an interdisciplinary approach involving the various levels of care would be appropriate. When we think about one patient group that has been identified as being particularly over-medicated, for example, the elderly, it is clear that social workers and other caregivers play important roles and probably see patients on a more daily basis and could be involved in the process of adverse event reporting.
Ms. Wiktorowicz: There is a role for practitioners other than physicians, who have limited time, as we have discussed. Maybe we could make that more explicit than it currently is, but we would need a different model of practising health care than we currently have, which not always has a role for different kinds of practitioners.
The Chair: This is an issue that came up in our study of the health accord, where we identified clearly that the silo approach is one of the big issues and it relates directly to your question, senator.
I would like to come to an issue that has not come up explicitly in the testimony today, but it has indirectly. You have talked about the question of the effectiveness of drugs in that effectiveness is also a potential adverse reaction if a drug is not working as effectively as it is presumed.
It is known that the composition of the material in which a drug is packaged has a great deal to do with how the drug performs when delivered into the body. I am trying to be as careful here as I can with this. There are reports that generic drugs, which often have a different composition of the environment in which the active ingredient resides from that of the patent medicine that was originally approved, have different levels of effectiveness in the body.
A very recent report came out last week of a serious issue in the United States where the FDA has made a ruling for the first time with regard to drugs that have long-term release impact. That is a current trend in a number of drug types — to take one pill and have it release slowly over a period of time. They report that the generic version has been identified as having no effect compared to the original patent medicine where the active ingredient in a different delivery form presumably had an effect, as established during the lifetime of the patent.
My question is now back, specifically: Once a drug goes off patent and goes into the generic realm, do you think there need to be more deliberate attempts to determine generic drug efficacy versus the original drug?
Dr. Baylis: I can offer two comments. It is my understanding that if a company comes up with a generic medicine that is now off patent, if they can show that the chemistry is the exact same molecule, they do not have to go through any kind of regulatory process at all. They just provide documents saying they are using exactly the same chemical compound, and they are absolved of any of the rigour that the first company would have gone through in terms of the market authorization.
My second comment has to do with your issue with respect to the composition of the packaging material. I do not know about that specific issue, but I can say that, in the early days when I worked with a clinic that did in vitro fertilization, all of a sudden they got no pregnancies. They had no explanation for why they had gone from being a very effective clinic in terms of getting pregnancies to having none. In the end, they came to the conclusion that they had bought new labels from a different company and they were labelling their materials with them. They changed back to the old company and the pregnancies came again. There was no proof — they did not do a randomized control trial afterwards — but it certainly leads me to have some sympathy to the claim that it is quite possible there are chemicals in things other than the actual active drug. It is probably worth looking into.
The Chair: Professor Lemmens, do you have any input?
Mr. Lemmens: I want to make one comment. I am not a pharmacology expert, but at the same time my reading of the literature is that these claims are often overstated. Very often those claims are made by the brand name industry to defend the sale of patented medicines and the instances where there is really a pharmacological difference are rare. I would say any studies that can come up with these arguments have to be treated with caution and must be analyzed as to the source of the study. Obviously, such claims have to be taken seriously, and you indicate some interesting examples that I was not aware of, but I know that claims are sometimes made too easily in that context.
The Chair: I think we can say for certain that the composition is important because that is part of the approval process in the early phase. However, you have made your points very well and I appreciate it.
I would like to finally come to an aspect that Professor Baylis touched on here and in previous appearances before us. I will not ask a question here, but I will pick up on the input in it because I think it ties into our ability to monitor drugs post-approval. It is the extreme importance of being able to have an effective way to monitor the performance of drugs post-approval to get at — in addition to all the things we have talked about, the specific issue that you raised and Senator Seidman raised in a number of our meetings — the population subgroups. This is obviously pregnant women, children and the elderly. We are now increasingly aware that when you take very large numbers of people or those with heart conditions there are many subgroups within any population.
It seems to me that we must really have a clear objective to try to find a way to do several things. First, we have to get input on adverse drug reactions and second, have them identified so that you can deal with the group that is being effected; not the individual necessarily, but the group. The post-approval monitoring would constitute a tremendously effective quasi-post approval clinical trial if we could get the data related to the specific subgroups.
In this regard, the issue comes back to the question that Senator Eggleton raised and that I followed up on, which is how we more effectively get the input of a potential adverse reaction. I will stick to adverse reactions now. You have made the points that not only are there adverse reactions but there is the actual effectiveness of the drug, whether it is doing anything, positive or not, let alone harming someone directly with an additional reaction. It could be harming them because it is not doing any good, so it is a complex area. However, I want to leave all three of you with a take away question because if you can use this as the example, then we may be able to tailor it with regard to the larger questions of drugs that do not have any effect or some other aspect.
Could you think about the issue of trying to get a higher degree of input of potential adverse reactions that would be easy for the patient, the doctor, whatever, and then any further thoughts you might have with regard to what the process might be in terms of analyzing and following up on those results once they come in.
I do not want an answer on that now. You have given aspects of that answer throughout the testimony today and we do not have time left to go into it in detail. However, I think our committee sees this as a very important aspect, potentially, of helping with many of the issues that surround the post-approval issues and that you have identified here today. We would certainly welcome your input on that.
I will go further and say that if there is any other issue that has come up during the course of our meeting, and you leave here thinking, "Darn, if I just thought to mention that," if you could get that to the clerk we would most appreciate it.
Finally, on behalf of the committee I want to thank you for appearing here today, the two of you in person here with us and Professor Lemmens for again appearing before us but taking extra effort today to do so. Your testimonies have all been extremely helpful to us and I want to thank my colleagues for the clarity of their questions.
With that, I declare the meeting adjourned.
(The committee adjourned.)