Proceedings of the Standing Senate Committee on
Social Affairs, Science and Technology
Issue 24 - Evidence - October 31, 2012
OTTAWA, Wednesday, October 31, 2012
The Standing Senate Committee on Social Affairs, Science and Technology met this day at 4:21 p.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 hereby call this meeting to order. The first thing I will do on this occasion is officially welcome our guests from the FDA in the United States. We have Dr. Gerald Dal Pan, Director, Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, United States Food and Drug Administration; and Dr. Robert Temple, Deputy Centre Director for Clinical Science, for the Center for Drug Evaluation and Research.
To our very distinguished guests, on behalf of our committee I want to say how privileged we consider ourselves have you both here with us today by video link. You have received background with regard to where we are in this study. This is the second phase of a four-part study we are doing on prescription pharmaceuticals. In this particular case, we are looking specifically at post-approval monitoring.
I will ask my colleagues at the table to introduce themselves. We will then invite you to make a presentation, which will be followed by questioning from the committee.
As usual on a Wednesday, I will ask my colleagues to introduce themselves.
Senator Eggleton: Art Eggleton. I am a senator from Toronto and the deputy chair of this committee.
Senator Munson: I am Jim Munson, and Ontario senator and the opposition whip in the Senate.
Senator Dyck: I am Senator Lillian Dyck from Saskatchewan.
[Translation]
Senator Verner: Senator Josée Verner from Quebec.
[English]
Senator Seth: I am Senator Asha Seth from Toronto, Ontario.
Senator Martin: I am Senator Yonah Martin from Vancouver, B.C., and the deputy government whip.
Senator Seidman: Senator Judith Seidman from Montreal, Quebec.
The Chair: I am Kelvin Ogilvie, a senator from Nova Scotia and chair of the committee.
On behalf of my colleagues, I invite you to make your presentations. I understand that Dr. Dal Pan will be presenting but that both of you are available for questioning as you determine appropriate. Please proceed.
Dr. Gerald Dal Pan, Director, Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, United States Food and Drug Administration: Thank you for that nice introduction. Good afternoon. I am Dr. Gerald Dal Pan, the Director of the Office of Surveillance and Epidemiology at the United States Food and Drug Administration's Center for Drug Evaluation and Research. I am joined today by my colleague, Dr. Robert Temple, who is the Deputy Centre Director for Clinical Science at the Center for Drug Evaluation and Research.
Dr. Temple and I are pleased to appear before the Standing Senate Committee on Social Affairs, Science and Technology to discuss post-marketing monitoring of medicines. I will make some brief opening remarks, after which Dr. Temple and I will be pleased to answer your questions.
Drug safety is a top priority at the FDA. Monitoring and understanding the safety of drug and therapeutic biologic products is a process that proceeds throughout the product's life cycle, spanning the period prior to the first administration to humans through the entire marketing life of the product.
Pre-approval drug safety assessment is an extensive process that involves preclinical safety assessments, such as animal toxicology testing, clinical pharmacology studies and clinical trials. Pre-approval clinical trials provide the efficacy and safety information that form the basis for an approval decision. The pre-approval safety assessment quantifies and characterizes the common adverse events associated with a medication. Depending on the number of subjects exposed prior to approval, less common adverse events can also be detected. The FDA reviews all the data collected during drug development to determine if the benefits of the drug outweigh its risks. If the benefits of the drug do not outweigh its risks, the drug is not approved.
Though pre-approval drug safety testing is a very comprehensive exercise and is rigorously reviewed prior to a drug's approval, no drug development program can identify all risks associated with a product. It is therefore imperative that drug safety assessments continue in the post-marketing period, when large numbers of persons will be exposed to the medicine, including many with coexisting medical conditions or on concomitant medications that were not represented in the pre-approval clinical trials. The goal of the post-marketing safety program is to identify adverse events that were not identified prior to approval and to understand better the spectrum of adverse events associated with the drug, including adverse events identified prior to approval.
Post-marketing risk assessment can be based on either observational data or clinical trial data. Observational data includes individual case reports of suspected adverse reactions, also called spontaneous reports, case series of such reports, analyses of databases of spontaneous reports, disease-based registries, drug-based registries, electronic medical records systems, administrative claims databases, drug utilization databases, poison control centre databases and other public health databases that track medication usage. The particular approach used depends on the specific drug safety issue. In certain cases, the FDA can require a company to perform drug safety testing post-marketing.
FDA staff carefully review post-marketing safety data to identify previously unrecognized risks or to characterize a known risk more fully. Some of these risks will be sufficiently serious to alter the benefit-risk balance of the medicine, such that post-marketing regulatory action will be needed. Possible regulatory actions include updates to the professional labelling, development of or updates to patient labelling, use of additional means of communicating risks to patients or professionals, introduction of specific risk management measures, restrictions on the use of the medicine or, rarely, market withdrawal. The FDA has the authority, in certain instances, to require a company to change its label for safety-related reasons or to institute a risk evaluation and mitigation strategy, a type of risk management plan.
Drug safety assessment requires the input of many disciplines, including clinical medicine, pharmacy, pharmacology, epidemiology, toxicology, genetics and many others. The sciences that underlie drug safety are evolving, and FDA scientists strive to incorporate the most advanced scientific thinking into their analyses. Our approach to drug safety assessment at the FDA is a multidisciplinary one that ensures all relevant expertise is brought to a drug safety issue. We apply the same degree of management to it that we apply to pre-approval market review.
Communication to the public is an important part of our post-market safety program as well. We have several programs in place to keep the public informed of our ongoing post-market safety monitoring.
In summary, the safety assessment of a medicine occurs continuously throughout its life cycle. A variety of data sources contribute to post-market safety assessment and many disciplines are involved in the analysis of safety data. Rigorous post-marketing safety analyses lead to post-marketing safety-related regulatory actions. Communication to the public is a critical piece of the post-marketing drug safety program.
Thank you for your attention. Dr. Temple and I are happy to answer your questions.
The Chair: Thank you very much for that presentation. I will now turn to my colleagues on the committee to begin our round of questioning.
Senator Eggleton: I note your commentary about safety assessment of a medication occurs continuously throughout its life cycle, which I think would be very important. You have authorities that I do not think your counterpart in this country has. You have, for example, the authority to have drug-safety testing done; you have authority with respect to the labelling. You have outlined in your remarks a number of other authorities that you have.
There was a suggestion a few years ago that Health Canada, which is our regulator, would get some similar authorities, but in our legislative process that did not proceed to a conclusion and has not been reintroduced since 2008. Health Canada says that they can work with companies to implement the changes. They could use persuasion, and they do; they could withdraw the approval, I suppose. They could also bring it to the attention of the public or the medical profession that there are difficulties with a certain prescription, so that can have quite an effect. However, you have authorities provided for in the FDA.
How often do you think this authority is important? Are you able to work these things out, in most cases, with the industry, or do they resist? Do you get into non-compliance or penalties, and what kind of penalties would you invoke? How valuable is it to have those specific authorities in cases such as post-authorization studies and label changes?
Dr. Dal Pan: Thank you for your question, senator. These authorities are relatively new. We acquired them when the Food and Drug Administration Amendments Act was passed in 2007. Prior to that, we did not have the authority to require a company to change its label for safety reasons, nor did we have the authority to mandate a post-approval study. We have found this authority useful. We can only invoke it when certain conditions are met. These conditions are set out in law, and we have to have new safety information that would allow us to require a company to either do a study, change its label or institute specific risk management measures.
We have used these quite a lot. We can invoke the authority to require a post-market study either at the time a drug is approved — and require it in the approval letter — or after a drug is approved, when we learn of new safety information. We have data on how many of these we use. We can send that to you if that would be of interest to you. I, unfortunately, do not have these numbers off the top of my head.
For safety-related label changes, that is also a new authority. We did not have that prior to 2007. We use that sometimes, although, for a lot of safety-related changes, we do not invoke it because we can work quite smoothly with a company.
Dr. Robert Temple, Deputy Center Director for Clinical Science, Center for Drug Evaluation and Research, United States Food and Drug Administration: It helps, for obvious reasons, to have a hammer at the end of your hand on some of these matters. Negotiation is often helpful. However, you can see the contrast between safety-related post-marketing studies and effectiveness-related post-marketing studies. We are allowed to discuss post-marketing studies and new uses. For example, sometimes we discover that a drug is being widely used for a purpose not in its labelling. It can be quite difficult to get the company to go out and study that. They do not always want to do it. When we can identify a safety concern, we can in fact require it. It does help us to get what we want, although negotiation often works too because companies often have an interest in getting the information. However, it is plainly useful.
Dr. Dal Pan: One area where we find the authority for safety-related label changes useful is when we want to change the label on a class of medicines — multiple medicines in the same class. It is a very efficient mechanism.
Dr. Temple: For example, we changed all of the labelling for antidepressants and anti-epileptic drugs to warn about suicidal thinking and behaviour. The ability to require some of those things was helpful, although some of the changes happened through negotiation before that. It can help; no question.
Senator Eggleton: Thank you for that answer.
Let me ask you one more question. One of our prominent newspapers, The Toronto Star, a newspaper in my city, screamed out a headline the other day that Health Canada was ignoring a lot of the complaints about adverse reactions to drugs. I do not know how accurate all of this is, but they were saying that people who were bringing attention to certain drug reactions were being ignored. This would be people in the population; I think they even quoted a doctor in one case. They are saying that there is not a quick enough reaction to these reports and that meanwhile people are at risk of very serious illness or even death from some of these drugs. As I say, I cannot vouch for how accurate this report is and whether it is distorted or exaggerated in any way.
Do you have any particular time frame on how you deal with adverse reaction information coming from the public? Nowadays they can get it to you a lot faster through social media, as opposed to writing the traditional letter. Do you have particular protocols in dealing with this and getting back to the public?
Dr. Dal Pan: I mentioned in my opening remarks that we now apply the same level of management to post- marketing safety issues as we have traditionally done for pre-market safety issues. As part of that, we have a prioritization system for post-market safety issues. When we identify a significant post-market safety issue, one that could result in death, one for which we think we might withdraw approval or put warnings or other restrictions on the product, we prioritize it and classify it as either priority or standard. For priority ones, we try to review the data within six months. Standard is within 10 months. We also have a category called emergency that would be done a lot more quickly.
In recent years, we have put in this kind of structure. Again, we have a public document that explains our tracking of these issues.
Dr. Temple: I think you were partly asking about what you do with a report of some terrible reaction. Do you always pass it on? Do you always believe it is true?
Of course, part of what Dr. Dal Pan was describing is a process in which we try to make a sensible judgment about the thing that has been reported. Spontaneous reports, which is what we are talking about here — reports to an adverse reaction reporting system — can come in for a lot of reasons, and the crucial question is to distinguish between whether the drug did it or whether it was part of ordinary life. Ordinary life has bad things in it. These are complicated judgments, and a newspaper article can stimulate a massive number of reports. You just have to be conscious of those things.
One of the great things that happened recently is that we now have access to a number of sources of data collected from health plans and other places so that we can, far better than we ever could before, actually see if something that appears to be going on in our spontaneous reports is actually occurring within these systems. We can check better than we have ever been able to do, and we are doing it more and more. This is just coming into play, so we are not there yet. However, it is on the way and will help a lot in dealing with what appear to be worrisome reports.
Senator Eggleton: Thank you, gentlemen, for your answers. Please send along any additional information that can help us in our work.
Senator Seidman: Last week, when industry representatives appeared before this committee, I asked them for their comments on how we can best monitor the effects of drugs used in vulnerable populations after the drugs have been released on the market. As you stated, the FDA has acquired new authorities, one of which is that it can require post- marketing studies or clinical trials during or after approval, and that the FDA holds the authority to — and I quote from your new document — "describe the study or trial to be conducted, including how the study or trial is to be done and the population and the indication.''
What I am hoping for is that you might comment on what this means in terms of post-market surveillance of vulnerable population subgroups such as pregnant women, children and seniors.
Dr. Dal Pan: We are very interested in our post-market safety surveillance about specific populations, be they children, seniors, pregnant women or others. We have certainly had examples where we have identified cardiac risks, for example, in our post-marketing safety surveillance. When we look back at the clinical trials that formed the basis of approval, we find that people with cardiovascular risk factors were not included.
The same might be true with psychiatric adverse reactions; people with underlying psychiatric disease might not be included in those clinical trials. Then we will work with the company to have tests in those populations, if we feel that is necessary.
With regard to children, several laws have been passed in the United States, such as the Best Pharmaceuticals for Children Act and the Pediatric Research Equity Act, which mandate certain pediatric safety reviews be done after certain milestones in a product's life. Our staff and Dr. Temple's staff work on these. They are presented to a pediatric advisory committee two to four times a year that meets to look at our analysis of data in children.
When we talk about post-market safety, we are quite interested in what population we might be interested in here.
Dr. Temple: It is basically a requirement of those two laws that, essentially, every drug will be tested in children, unless there is some reason not to or the disease does not occur in children. Those things are done pretty uniformly
The Best Pharmaceuticals for Children Act provided an interesting incentive to make those studies happen. The company gets six months of additional exclusivity. They get an extension to their patent or other exclusivity if they do the trials that were requested of them. Between the virtuous nature of doing the trials and the benefits to the companies, they are getting done.
Senator Seth: My question is for Dr. Dal Pan. Does the FDA transfer into a global agency? If not, are they thinking of engaging with global agencies? If so, what efforts include placing FDA offices in manufacturing countries and harmonizing policies? What efforts are being made to increase the coordination between Health Canada and the USFDA?
Dr. Dal Pan: Drug regulation and drug safety is a global issue. We participate in a variety of global efforts. One of them that is relevant for post-market drug safety is the International Conference on Harmonisation. I believe Health Canada has representatives there as well.
The goal of the harmonization is that regulators around the world receive global data, and they receive the same global data so that they can talk with each other if a certain issue comes up where that kind of contact would be made.
We have regular teleconferences with Health Canada, though not strictly related to the topic here. We have also worked closely with Health Canada on medication and prevention as well.
Neither Dr. Temple nor I work much in the manufacturing arena. We both understand how important this is, but we are probably not the best people to talk to about that.
Dr. Temple: We exchange things with Health Canada often. Actually, a fair number of Canadian investigators carry out major trials and are very active, so we are in reasonably close contact about things and meet on drugs that cause difficult problems, things like that, as Dr. Dal Pan said. We are not quite all one system, but there are a lot more similarities than differences, actually.
Senator Seth: The last time we had CMA here, the doctor who appeared as a witness said that without an electronic system, we are not notified about adverse drug reactions and we cannot sufficiently provide knowledge to Health Canada or to patients. We do not have a regulated method here. We do not yet have implementing laws on these things. How do we continue to coordinate between Canada and the U.S.? How do we accomplish that? I do not quite understand.
The Chair: Dr. Dal Pan, I think there are two different questions there, one that I intend to come back to a little later, which is the issue of acquisition of the information from sources where adverse reactions may be potentially identified and how those get reported.
I think the end of the senator's question dealt with if in fact there is not a lot of electronic communication, how do Health Canada and the FDA communicate with regard to issues of mutual interest?
Dr. Dal Pan: There are a few ways. One is we have periodic teleconferences to discuss adverse events of interest with Health Canada and some other regulators. Those are on a periodic basis, probably bimonthly or quarterly. I am not sure of the exact frequency.
The other thing is, a lot of our safety-related issues are publicly on the web, and we can share some of that with those regulators with whom we have confidentiality agreements.
What we have found most useful is when there is a specific issue that we are interested in, then it is actually helpful to discuss that on an ad hoc basis. My own experience is that ad hoc work is actually very productive on a specific issue.
The Chair: Thank you.
Senator Seth: Thank you.
Senator Martin: I am curious about your final statement, which I think is very important with regard to what we are looking at with the overall post-market safety regime, communication to the public being a critical piece to your post- marketing drug safety program. I want to hear more about the several programs you mentioned that are in place and that are effective in your communication strategy.
Dr. Dal Pan: We have put a lot of effort into this in the last five to seven years. One of the things we learned 10 years ago was that people wanted more information, so we have a number of programs, many of which were initiated by the FDA and some of which are in legislation.
Probably the biggest one initiated by the FDA was a program of drug safety communications. It is a program through which we post on our website information about what we call emerging drug safety issues. These are issues where we have some preliminary information, we think there might be a problem, but we are not 100 per cent sure what the answer is. We put up what we have and we say, "This is what we are thinking of, this is the issue we are addressing, so stay tuned and we will update you when we have more information.'' This is a very active program. There is a website dedicated to it, and we can send you the link for it.
It has taken a lot of effort. This has gone through a lot of evolution over the years. We started with different communications for the public and for health care professionals. Now we wrap it into one single communication with tiered levels of information. What is unique about this program, what is different about it, is that we are telling people what we are concerned about before we have the final answer.
We have a few other programs.
Dr. Temple: Let me just add something. You need to recognize that this could be considered controversial. Some people would say, "Do not scare people until you really know.'' However, after a lot of discussion, we concluded that what people want to know is what is on our minds even before we are absolutely sure. It is an important decision, not without some controversy, but we obviously think it was the right thing to do.
Dr. Dal Pan: I agree with Dr. Temple. Our communications staff is really good, and they have walked us through improving this program to meet those goals.
In terms of other things we do, we have a large adverse event reporting system database. Under the Food and Drug Administration Amendments Act of 2007, which I mentioned earlier, we are required to screen that on a biweekly basis and put up all signals or potential signals of adverse reactions that we identify. On a quarterly basis, we post on our website signals of adverse events that we may be working on. Again, these do not necessarily represent our conclusion that the drug is associated with the adverse event, but it lets the public know we are working on it. This is updated every quarter to say what our conclusion might have been.
Another mandate from the 2007 law was for us to look at every newly approved medicine, and 18 months after it was approved or after 10,000 patients have taken it, whichever is later, we are to do a review of the safety of that medicine and post the findings on our website. This is another program we have done, and we communicate those findings as well.
Those are probably the biggest programs. We have put a lot of effort into these programs, especially the drug safety communication initiative.
We are also concerned about patient labelling. This is information given to patients when they pick up a prescription. In the United States now, most information that patients receive when they pick up a prescription in the pharmacy is actually not produced by the company and not regulated by the FDA. Under law, we were required to evaluate that. We determined this was not an effective system, and we are currently working on an improved patient medication information system. That is still in development.
Senator Martin: Thank you for those examples. I am thinking that there is a certain level of readiness you would need from the public to be able to look at information and have it be useful.
You talk about having tiered levels of information as well as having made this decision that may be considered controversial. Was there a period of trial and error where you had certain information but, based on reaction from the public, it was toned down? Do you now have these models that seem to work, and are other jurisdictions interested in perhaps using these same communication strategies? In what way could Canada look at some of these programs, and would they be available? There are a few questions there for you.
Dr. Dal Pan: First about the evolution. I mentioned that this program has evolved. When we first started several years ago, we had, as I said, one sheet for patients and one sheet for health care professionals. There were many different sheets out there and it was a little confusing. The biggest thing we did — and I credit our communications staff with this — was to put it all into one drug safety communication that is appropriate for all readerships.
In terms of toning things down, I think the balance has been more or less the same. We have worked with outside groups to explain to them what these are and what they are not. The web pages where they are posted actually describe the programs and the FDA's intent there. I should say that we also have a drug safety page on our website where all these items can be found, and we are happy to send you all the links.
Dr. Temple: One of the things that sometimes happens when you do this is that you have to say, after another year, "That was not real.'' However, we have decided it is worth letting people know what is on our minds, even if we have to do that sometimes, and it is better.
I want to mention this 18-month review. We are looking at the adverse events that come in all the time, and reviewing them. You might wonder why we need a special focused review after 18 months. The fact is that having a whole bunch of people in the room, all of them looking at this stuff together, sometimes pulls out a finding that individuals had not noticed before. It is a very healthy exercise. Usually we do not find anything major, but every once in a while we find something that was not described well enough in labelling or that calls for further study or something like that. It is really a very good thing to do, but it is not an obvious thing. That is why I mention it.
Senator Eaton: Thank you, gentlemen. This is very interesting. I will ask you a two-part question.
Do you struggle with the under-reporting of adverse events in the United States? What percentage of adverse events comes from producers, physicians, consumers, pharmacists and health care professionals? Could you give me ballpark figures?
Dr. Dal Pan: Yes, there is under-reporting of adverse events. This is widely described and widely recognized. In the United States, there is basically no requirement for a health care professional or a patient to report an adverse event to the FDA. There is widespread under-reporting.
The degree of that is talked about. You will often hear that 1 to 10 per cent of adverse events are reported. These are data from the 1980s, based on two surveys in states in the Northeastern United States. The actual percentage we get is probably not well known and it probably changes over time.
Senator Eaton: Do you have a definition for an adverse effect? Is there such a thing as a definition for an adverse effect?
Dr. Dal Pan: Yes. These are described in the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use guidelines. I do not have them in front of me, but unintended or noxious consequence of a medication, and it goes on to characterize that a little bit more. We do work from the ICH definitions of "adverse event.'' Our regulations have some definitions as well, as do our guidelines, but they are all the same, basically. There is widespread under-reporting of adverse events.
You asked also about the source of those adverse events. We accept adverse event reports directly from the public as well as from industry. About 95 per cent of our reports come from industry and about 5 per cent come directly from the public. This is a percentage that has been quite stable over time.
All adverse event reports that come from industry had to come from someone else first. They either came from a patient, a physician, a pharmacist or someone else. Our website does contain a lot of that information, and we can obtain that for you. We find that it is a variety of people. It is patients and consumers themselves, physicians, and pharmacists. I would say it is not one group that dominates.
Dr. Temple: It is worth mentioning that it is not under-reporting itself that is troublesome; it is the variable rate of reporting. If you get a report of something and you knew that another drug was reporting at the same rate and that you were getting 10 per cent of all of them, you could compare the two and make some sense out of it. What you do not know is whether one of them is reporting 1 per cent and the other is reporting 10 per cent, and there is no good way to know.
We know some things. We know that drugs that have been around a long time have lower reporting rates. That has been shown. However, it makes it very hard to know whether you have a finding or not, which is one of the reasons these systems that report certain kinds of events within a health care system are being looked at to get you better information. If heart attacks are captured by Kaiser Permanente all the time, then you will know how many heart attacks there are on people who got this or that drug and you can make a sensible comparison. If all you have is that some people sent some in and some did not, it is very hard to know what you have.
Senator Dyck: I will follow up on the questions that Senator Eaton posed. In fact, she asked the questions that I was going to ask.
You mentioned that you do have a definition of adverse drug reactions, but last week we had witnesses here from the Canadian Medical Association who indicated that sometimes it is not clear because there may be a reaction that occurs in a patient that does not seem severe, but it may turn out to be more significant later in time. How do you keep track of those?
Senator Eaton also mentioned that part of the difficulty is getting all the data together, because different health communities have different ways of collecting their data and the different computer network systems may not be compatible with each other. Are you confident that the data you are collecting is collecting all the data that is out there?
Dr. Dal Pan: Let me just make one distinction. You used the term "adverse drug reaction.'' A previous senator used the term "adverse event.'' They are slightly different things. An adverse event, according to the definitions we use, is any unintended or noxious consequence related to the medicine, whether or not it is due to the medicine. An adverse drug reaction is an unintended or noxious consequence that is due to the medicine. Part of our job is to separate out, as Dr. Temple said, what is due to the medicine and what is not.
To get into your question as to the confidence in the data and knowing what really happened, spontaneous reports, or these reports that people send in, are only as good as the information they put into them. It is generally a blank narrative text field where they are free to write as much or as little as they want. Someone might give a lot of clinical details; other people might give very few. We have a requirement that when the adverse event is serious and it meets a regulatory definition and is unexpected, meaning it does not occur in the product label, that pharmaceutical companies, upon receipt of such a report, have to perform follow-up to get as much information as they can. On occasion, FDA staff will also do follow-up on reports, but we are not staffed to do follow-up on all the reports we get. When there are serious safety issues identified and we are interested in a set number of reports, we will try to get as much follow-up as we can.
I need to emphasize that there is no requirement to submit the report to FDA or to the company. There is no requirement to provide the follow-up. The quality of the reports is actually quite variable.
When we look into electronic databases, most of the data there are based on coded bio-informatic terms, and the utility of those findings from the database are only as good as the correlation between the coded term, which is used for billing in the United States, not strictly for the delivery of medical care. If the issue is really important, we will actually do a validation study comparing the billing codes to the actual medical chart. That is something that is done only rarely because it is time-consuming and expensive.
Senator Dyck: With regard to the under-reporting of adverse drug reactions, is there anything you have done that maybe has optimized adverse drug reaction reporting or suggestions on ways that the reporting from the public could be improved?
Dr. Dal Pan: My own view is that what is really important is the quality of the reports, not so much the number. What we really want is good-quality reports, and good-quality reports do not have to be very long. Some of our staff have attended medical meetings, worked with professional societies, given talks and things to try to improve that, but there has been no system-wide effort to do that. We try to raise awareness, though.
Dr. Temple: Every package insert, every drug label, now has a phone number you can call in the highlight section, telling you where to phone in your adverse reaction reports. We are trying to make people more conscious.
For what it is worth, I go back a long way here. Our spontaneous report system used to get about 20,000 reports a year. In the most recent years we are up to 600,000.
Dr. Dal Pan: It was about 900,000 last year.
Dr. Temple: Whether it is us or something else, something has made people much more interested in telling us about these things.
The Chair: Gentlemen, I want to pursue this last area a little further. Clearly, if one is to be able to evaluate the adverse events around particular products, one has to get a report of some sort to some centre that recognizes it and is in a position to collect and then perhaps carry out some action as a result of them. However, if the reports do not come in, you can have all kinds of adverse events, but we do not know about it in a way in which a regulator can perhaps help protect the public. You have made comments in that area in general.
With regard to reporting, I would like to ask you some specific follow-up questions to further clarify your answers to my colleagues' earlier questions.
You indicated that you have the ability to track the number of prescriptions. You indicated, if I recall correctly, that after 10,000 prescriptions or 18 months you do an automatic review of a given pharmaceutical. In order to get 10,000 prescriptions, you have to have some way of knowing how many prescriptions are out there. Is that reported by the various drug plans that are in existence or is there some other mechanism? Is it collection through pharmacies? How do you ensure that you collect the number of prescriptions for a particular product?
Dr. Dal Pan: We actually purchase data — we call it drug utilization data — that tells us how many prescriptions were dispensed in a given time period for a given product. That is commercially available and we purchase it.
The Chair: Who collects it so that you can buy it from them? What is the process by which the prescriptions are collected?
Dr. Dal Pan: For outpatients, it is generally linked to billing systems. We have a complex health insurance system here in the United States whereby when you go to a pharmacy, they have to send something to the billing organization. It is basically billing that drives these systems. We can get you more information on how that works. The system of collecting this information is not an FDA system. For outpatient prescriptions, it is mainly linked to billing records collected by third parties that work with the pharmacies and the chains to amass this data.
The Chair: Would IMS Health be involved in that?
Dr. Dal Pan: That is one of them, yes.
The Chair: Thank you very much. That is very helpful.
You have made very important points here that it is the quality of the report that is essential to the regulator or, indeed, the producer, to be able to evaluate what is happening with the actual product in the market. We have seen a number of reports in the literature that the reporting level of adverse events may be as low as 5 per cent of the total. I do not want to get into a question of the validity of that data, but all reports indicate that a low percentage of the actual number of adverse events get reported.
Before I ask the question, you also indicated that you do not have the staff to carry out a full review of each event as it is reported to you or you become aware of it. The question I have for you is the following: With regard to pharmaceuticals in the marketplace, is there not a point at which it is important to make it easy for the patient, nurse, doctor or pharmacist to send a signal that there are signs of adverse events occurring, such that those can be collected and someone will have the ability to follow up in such a way that there is a faster awareness of potential problems with a given product?
Dr. Dal Pan: Let me back up a bit. With our adverse event reporting system, the reports we receive are quite important to us. They account for probably about half of the drug safety communications and safety-related label changes. This is not the only source of drug safety information. There are some adverse events for which the reports themselves are not going to be very valuable, even if they are quite complete, in determining if an adverse reaction is really related to a drug or if it is related to something else. You need other kinds of data for that.
An example would be the occurrence of a heart attack in patients on diabetes drugs. Diabetics are at risk for heart attack, so one report will not be sufficient for inferential value, nor would a collection of them. You need more population-based data with proper numerators, denominators and comparators.
Let me get to your question about making reporting simple and easy. We would love a system that would make it simple and easy for physicians, nurses, pharmacists and people at the point of care who have the best knowledge of the patient's condition and the possible role that the drug could have had in the development of an adverse event.
Currently, we rely on them writing a narrative. We recognize that takes time and that these clinicians are quite busy. People have talked about using the electronic medical records system to signal adverse events related to medicines and that could automatically generate a report.
A study was done by a group in Boston that used an electronic medical records system to send reports of suspected adverse drug reactions to the FDA. That is published. We could send it to you. We also did an analysis of that and published it. We learned a lot about something we already knew about, not about new things. We also found that some of the electronic data was not the most up-to-date. It is important that you have the best possible data, and if it could be automated in some way, that would be great.
The Chair: The point you are making around the quality of the report and the complexity of these issues is exceedingly important and we are not trying in any way to trivialize that. On the other hand, unless reports start coming in some way, one does not drill down to the kind of evidence that is needed to make the conclusions that you are talking about. We are trying to find ways that might encourage reporting that would be helpful in that area.
To refer to one specific study that occurred here in Canada in a pediatric area, there was an attempt over about an eight-year period to encourage health professionals and patients to be aware of reporting and the importance of reporting. A deliberate effort was made to encourage people to be aware of this. It appeared to have some positive effect in that particular focus area.
You alluded to certain situations in the U.S. in some of your testimony. Are you aware of any specific efforts that have been deliberately undertaken to inform health care professionals and patients of the importance of reporting and making them aware that this is something to do? You mentioned some specific examples, but I would like you to focus your answer specifically on that issue.
Dr. Dal Pan: There have been small-scale efforts to encourage reporting. I think there was something published several years ago. We can try to get that for you. I am not aware of any national-level effort directed at reporting adverse events.
Dr. Temple: Nor am I. You see attempts to get reports on reactions to a particular drug, generally stimulated by attorneys, but of course that gives you an unbalanced picture of what all the drugs do. You are asking about whether we could get more birth defect reporting or something. I am not aware of anything like that. I would say our general view is that having more birth defect registries is the way to do that because then you get all of them.
Dr. Dal Pan: Dr. Temple referred to stimulated reporting. When there is widespread attention to a drug safety issue in all the major national newspapers and news shows, the numbers go up and come right back down. However, that is an after-the-fact type of event. We have already made a conclusion at that point generally, and the reports are just stimulated. I share your interest in getting better reports earlier, especially for some of the rare reactions.
The Chair: Your points are extremely well made and we certainly understand that. The effort I was referring to was one that was not stated in the national newspapers but was one of those things where the pediatric society worked with the health professionals and they felt it had a positive effect in that area.
I want to shift for a moment to one group within the total health care system: the pharmacists. In another study by this committee, we were informed that the pharmacists often have a lot of information with regard to the use and impact of particular drugs. As a simple example, in a flu season, the pharmacists are often the first ones to know that flu is active in a given area because the number of prescriptions goes up fairly dramatically in this area. They are quick to recognize that something is going on. The Canadian Pharmacists Association also indicated that they have a fairly sophisticated IT system, data collection system, and might be more useful in general. Do you have any comments with regard to the role of pharmacists specifically in this area of reporting adverse events?
Dr. Dal Pan: Pharmacists do report adverse events and have historically been responsible for a reasonable percentage of the adverse event reports we receive. Yes, pharmacists have a role. We will send you the breakdown of pharmacists, physicians, et cetera. Pharmacists are certainly knowledgeable about the effects of medicine, the potential for drug interactions, safe use conditions of the medicine, and are in a position to counsel patients on that. Yes, pharmacists do report adverse events to the FDA.
The Chair: My final question in this area is related to an answer that you gave earlier with regard to the collection of information as it relates to groups that are often under-represented in the initial clinical trial stage. Pregnant women, children and the elderly were specifically mentioned. Could you elaborate a little bit on your answer? I thought your answer was extremely important to us with regard to how you have been able to gather data for these groups post- market approval, and that this represents extremely important data with regard to further prescribing to previously under-represented groups.
Dr. Dal Pan: Let me talk first about an initiative we have undertaken to understand prescribing practices to pregnant women and a resource we have been developing to understand the effects of a medication used in pregnancy. We have access to certain contracted data where we can contract with outside experts, generally university experts or health plan experts, who have access to data and who have the expertise to work with us on those data. We developed a system whereby we can identify drug usage in pregnancy and we have been able to characterize, for certain common medicines, the frequency of drug use in pregnancy. What is unique about this is we are able to link it to the baby's outcome after the pregnancy. We are undertaking some studies to look at the effects of medication in pregnancy.
As Dr. Temple mentioned, when we are concerned about a certain risk of a medicine in pregnancy, we can ask the company to set up a registry, so they can follow women taking the medicine who are pregnant and look at the outcome of the baby, or there are some established registries that companies can work with to get some of this information.
Those are the efforts we have for pregnancy-related medication exposure.
We spoke earlier about the initiatives we have for pediatrics, these systematic reviews of medicines used in children that occur under two different laws we have.
We are also interested in adverse effects of medicine in the elderly. The elderly take quite a lot of medicine. They are much more likely to be on multiple medications than younger people. We have government-sponsored health insurance in this country for people over 65 called the Medicare program, run by one of our sister agencies in the Department of Health and Human Services, and we have a relationship so that we can use their data that links medical diagnoses to prescriptions and study drug effects in persons over 65. If we have specific questions, we can query their database on that. That is probably the best resource we have for the over-65 age group.
The Chair: Dr. Temple, you referred to something in one of your answers that is obviously extremely important in terms of selective evaluating of adverse events. In order to make any comparison of the impact of the various medications, it is essential to know that you are getting roughly the same rate of reporting for each of the medications. Based on the other testimony today, it is obviously not possible to make absolute assurance that that will occur. As we get closer and closer to personalized medicine that relates to various subgroups of the population, for example, medications that relate to heart disease — we know there are many subpopulations within that particular group, and it is often important to be able to distinguish the impact of a drug on the subgroups — it would be wonderful if we could find a way to achieve the solution to the problem you identified so that the information being collected through post-market awareness of adverse events might very well be helpful to longer-term prescribing for genetic subgroups within a large population.
Do you have any general comment to make beyond your earlier response?
Dr. Temple: I spend a lot of my time travelling around to places telling them to keep their hopes under control about the possibility of doing this in post-marketing surveillance systems. It is important to remember that differences between drugs within a class, or at least across a class, or even across two classes, are likely to be modest most of the time and that, unfortunately, the way you are going to discover important differences is through properly done randomized trials. Having said that, every once in a while something turns up that suggests a large difference and these systems might be helpful on that.
I will give you an example, but I cannot give you the results yet because they are not available.
When a new anticoagulant called dabigatran was approved, we got a large number of reports of bleeding, far more than we got for its predecessor drug, warfarin or Coumadin. As Dr. Dal Pan described before, we put out a drug safety communication to let people know we were careful to say we were not sure this was a true bill. We did not know. The reason we did not know is a large, 15,000-patient controlled trial had showed no difference in bleeding. However, you cannot ignore these results. They might mean something we were not smart enough to figure out in the controlled trial. We are using some of these epidemiologic systems to look at bleeding rates within these systems on the two drugs, and, not too long from now, we will have an answer on that question, but I cannot tell you anything about it yet.
That tells you that if there seems to be a very large difference in the spontaneous reporting system, these systems might be able to verify it or tell you it does not really look that way. For big differences, these systems will be very good. We are planning to use them just that way, to take these signals of something that really looks bad and see if we can see it in a more controlled setting. That is not going to tell us which antidepressant is better, however.
The Chair: I am not surprised by your answer, but I wanted to hear it from you who have so much expertise and experience in this area.
On behalf of my colleagues, I want to express our deep appreciation for your appearance before us today, and most emphatically for the breadth of knowledge you have been able to bring to the questions that have been asked, and the experience, of course, that your country, with a much larger population, has with regard to many of these important issues.
As you have also indicated, this is a very important area, but the systems in which we can identify, collect and then properly evaluate adverse events is still very much in its infancy in terms of getting a high percentage of reports of adverse events.
Colleagues, I want to thank you for your questions.
(The committee adjourned.)