Health Committee on April 10th, 2008

[The Vice-Chair, Liberal, Lui Temelkovski]

I call the meeting to order.

Pursuant to Standing Order 108(2), this is a study of the post-market surveillance of pharmaceutical products.

We have a number of witnesses with us today. We have the Auditor General, Sheila Fraser, and Neil Maxwell from the Office of the Auditor General; as well, from the Office of the Privacy Commissioner of Canada, Jennifer Stoddart, the Privacy Commissioner herself, as well as Patricia Kosseim, general counsel. From Canada Health Infoway, we have Richard Alvarez, president and chief executive officer, as well as Mike Sheridan, chief operating officer. Welcome.

I think we all know that each group has 10 minutes to make a statement, and then questioning starts thereafter.

We'll start with Madame Fraser.

[Sheila Fraser, Auditor General of Canada, Office of the Auditor General of Canada]

Thank you, Mr. Chair.

We thank you for the invitation to participate in your study on post-market surveillance of pharmaceutical products. As you mentioned, I'm accompanied by Neil Maxwell, assistant auditor general. I believe that some of our past findings on regulatory and post-market surveillance activities at Health Canada would be useful to the committee in its study.

In my presentation, I will highlight some of our findings on the drug products program discussed in our 2006 chapter entitled “Allocating Funds to Regulatory Programs--Health Canada”. I am also pleased to talk about two other audits, where we examined programs similar to pharmaceutical products. We reported these findings in our 2004 chapter on regulation of medical devices and our 2000 chapter on the regulatory regime of biologics.

I will start with our 2006 Chapter—Allocating Funds to Regulatory Programs. In this audit, we examined issues related to the allocation of financial resources in three regulatory programs.

These programs regulate the safety and use of consumer products, medical devices and, what is of more interest today, pharmaceutical products. It is important to note that the audit was on the allocation of financial resources and how this affected that Department's ability to carry out its regulatory responsibilities.

We did not examine other aspects of pharmaceutical products, such as post-market activities. We concluded that Health Canada did not know if it was fully meeting its regulatory responsibilities for the Drug Products program.

First, Health Canada needed to determine the activities that must be carried out in order to meet the Department's regulatory responsibilities. Program managers for Drug Products indicated that they considered the level of post-market compliance and enforcement activities to be insufficient. This could have consequences for the health and safety of Canadians, such as exposure to unsafe, ineffective or dangerous products.

Second, we found that Health Canada needed to determine performance targets for the activities.

Third, Health Canada needed to determine the level of financial resources required to carry out the activities necessary to meet is regulatory responsibilities. We found that the demands on regulatory programs were increasing while the funding remained constant, making it difficult for program managers to fully meet the Department's regulatory responsibilities. Although the funding remained constant, the core funding for the Drug Products program had decreased 32% over three years, if all sources are considered.

Furthermore, most of the additional funding that Health Canada received had been allocated to pre-market activities, and funds that were meant for the Drugs Product program were reallocated to other programs. For more details about funding, please refer to the table we distributed.

In its responses to our recommendations, Health Canada stated that it planned, among other things, to improve the operational planning process; review the funding, including core funding allocated to regulatory programs; work on a cost recovery strategy and regime, including the establishment of a full costing model; introduce a budget management framework with guidelines on resource allocation and monitoring of Treasury Board decisions; and improve performance measurement.

Health Canada also promised to conduct comprehensive reviews that would serve to establish program baselines by defining the required level of activities that the program must carry out, identify the targeted performance for these activities, and identify the resources needed to carry out these activities.

Health Canada promised to make these changes by March 31, 2008, or earlier. The department's next progress report is scheduled for the end of April, and your committee might wish to examine what progress has been made, especially concerning the baseline study for pharmaceuticals.

The two other audits were about programs similar to pharmaceuticals. We found similarities in the findings related to post-market activities. We would ask you to note that these audits are dated, and we have not conducted recent audits to see if our concerns have been resolved.

Post-market activities were a concern in our 2004 audit on the regulation of medical devices. We found that health Canada did not carry out any inspection activity at the post-market phase and did not know the extent to which the regulations were being respected.

The Department had made it mandatory for the manufacturers to report serious adverse events, but had done little work to increase the number and quality of reports received from health care professionals. The rate of reporting on adverse events was significantly less for Canada compared with the US and the UK. In addition, there were weaknesses in the analysis and interpretation of adverse events, and there was no proactive system to identify patterns that could signal a serious safety concern.

We also noted that Health Canada had never developed a communications plan or strategy for medical devices to ensure that Canadians were fully aware of risks.

In addition to recommending that Health Canada address these weaknesses in post-market activities, we also recommended, due to the weaknesses also found in pre-market activities, that Health Canada make a choice: either provide adequate human and financial resources to deliver the program as designed, or redesign the Medical Devices program and the Regulations to manage risks in a way that requires fewer resources.

Finally, in our 2000 audit on the regulatory regime for biologics, we found that Health Canada had difficulty managing the workload of pre-market and post-market activities. Officials told us they had trouble staffing positions. We recommended that sufficient databases be implemented to adequately process, analyze, and disseminate information on adverse reactions and events for biologics.

As you can see from these three chapters, we have had concerns about resource allocation and post-market surveillance for several years now. It seems that the emphasis has been on pre-market activities, to the detriment of post-market activities.

As you know, I cannot comment on the policy approach, but I am encouraged to see that some of the issues raised in our reports, such as problems with post-market surveillance and under-reporting of adverse events, are included in the government's Blueprint for Renewal: Transforming Canada's Approach to Regulating Health Products and Food.

I would like to take this opportunity to update the committee on other work we are conducting. We have a chapter on infectious disease surveillance in our report to be tabled on May 6. We would be pleased to meet with you again after tabling to discuss that.

We are also beginning an audit on electronic health records, and we plan to provide this report to Parliament in fall 2009. Several provincial auditors general will concurrently carry out audits on the same subject matter, leading to a comprehensive look at the implementation of electronic health records in Canada.

Lastly, we will soon begin developing a plan identifying the audits we will conduct over the next five years. We would be pleased to discuss our plan with you at a future date.

I hope our comments today will help your study on post-market surveillance of pharmaceutical products. I look forward to reading your report and the government's subsequent answer to it.

Mr. Chair, that concludes our opening statement. We would be pleased to answer any questions the committee members may have.

Thank you.

[The Vice-Chair, Liberal, Lui Temelkovski]

Thank you very much, and thank you for staying within the time. It sounds like you've done this before.

We'll continue with Madame Stoddart.

[Jennifer Stoddart, Privacy Commissioner, Office of the Privacy Commissioner of Canada]

Thank you very much, Mr. Chair.

I am very pleased to speak to the committee about the privacy implications of the post-market surveillance of pharmaceutical products. With me today is Patricia Kosseim, General Counsel and an expert in health law.

You have received a fairly comprehensive document prepared by our Office which starts out by saying that while Canadians regard the health care they receive as a top priority, they also consider ongoing privacy protection to be very important.

This morning, I will begin by briefly discussing some issues that are addressed at greater length in my submission, a copy of which has been circulated to members. These include the potential re-identifiability of data, the privacy implications of electronic health records, data breach notification requirements and finally, the concept of “work product” information.

The Privacy Act applies to federal government institutions, agencies and Crown corporations. As such, it applies to government health surveillance programs such as Health Canada's Canadian Adverse Drug Reaction Information System or CADRIS, and other government initiatives, such as the Federal Health Care Partnership's plans to develop electronic health records.

I'll start with the issue of re-identification of data as a privacy issue in post-surveillance of medications.

From a privacy point of view, one of the key issues we grapple with is the concept of re-identifiability, particularly in the era of increased digitization of health data and surveillance programs, proliferation of publicly available information through the Internet, and sophisticated technological capacity to link up information across different databases. Personal information is critically defined in both the public and private sector law as “information about an identifiable individual”. Exactly what is identifiable or potentially identifiable is a relevant issue for your present study.

Re-identification was at the heart of a recent decision in January of this year by the Federal Court in the matter of Gordon and Health Canada and the Privacy Commissioner of Canada. We were interveners. I bring to the attention of this committee four points.

First, in a situation involving personal information about an individual, the right to privacy is paramount over the right of access to information. That was the first major conclusion of this recent finding, which is not being appealed.

Second, the Federal Court adopted the legal test that was proposed by my office, and I quote: “Information will be about an identifiable individual where there is a serious possibility that an individual could be identified through the use of that information, alone or in combination with other available information.”

Third, the court concluded that disclosure of some information, in particular factual circumstances, where it is combined with personal information, is to be scrutinized for its effect on personal information.

Finally, the Federal Court emphasized the importance of ministerial discretion in deciding whether or not to exceptionally release this personal information in the public interest.

On privacy considerations in electronic health records, major initiatives under way to develop electronic health records promise great things for Canada's health care system: improved quality, efficiency, productivity of health care services, enhanced patient safety, more evidence-based decision-making, facilitated knowledge transfer, and greater accessibility to services and treatment.

So as health information structures proliferate across the country, the traditional lines between health care, surveillance, quality assurance, and research will become increasingly blurred. This is not necessarily a bad thing; however, the notion of purpose, which is such a critical concept in data protection laws, and the ones individuals actively turn their minds to when they provide informed consent in any meaningful way--we all ask what you want this information for, and what's going to be done with it--is increasingly being challenged by this approach.

As the concept of purpose becomes stretched, other purposes can begin to creep in. Beyond health-related purposes are other more worrisome purposes to which personal health information may eventually be put, particularly as external pressures for such information continue to rise. Marketing, employment, insurance considerations, law enforcement, and national security are just some purposes that loom on the horizon. These are clearly not part of the deal that Canadians think they are getting themselves into when they think of the development of electronic health records.

Another critical concept that is increasingly being challenged in the context of EHRs and electronic clinical trials is the central concept of accountability, particularly as more and more entities join up through interoperable systems, as public-private partnerships develop to leverage resources and achieve commercialization objects, and as data flows across provincial and national borders in a global economy. So I guess that's where I join up with my colleague the Auditor General.

In order to help work through some of these challenges, our office is participating in the recently created Canada Health Infoway privacy forum that brings together representatives of the health ministries and privacy oversight offices across Canada. We're very pleased to be part of this critical discussion that is starting to address issues of informed consent, secondary purposes, and accountability as they relate to the implementation of interoperable pan-Canadian electronic health record systems.

The third issue is data breach notification requirements as they relate to privacy.

With the growing digitalization of health data also comes increased scope and impact of potential breaches. A number of recent cases which I highlight in my submission have brought this problem to light. Not a day goes by in Canada without a report of someone finding identifiable personal health records in a trash cash behind a clinic, hospital or doctor's office.

Industry Canada is currently looking at how to incorporate into the Personal Information Protection and Electronic Documents Act, or PIPEDA, mandatory breach notification requirements. This is a welcome development which we hope will serve as an incentive for organizations to put proper security safeguards in place and to be open and transparent when something goes wrong.

In the meantime, our Office has issued guidelines to support organizations through critical actions steps, including assessing the risk and extent of potential harm, and deciding when, how, who and whether to notify individuals. When dealing with highly sensitive personal health information, special considerations should be taken into account, such as psychological risk of harm.

I would now like to turn to the section of my submission on work product, an issue that was discussed at length when other parliamentary studies were conducted on personal information protection. I am available to answer all of your questions about how concerns over protecting information apply to this area.

[The Vice-Chair, Liberal, Lui Temelkovski]

Thank you very much, Madame Stoddart.

We'll continue now with Mr. Richard Alvarez.

[Richard Alvarez, President and Chief Executive Officer, Canada Health Infoway]

Bonjour, Mr. Chair and members of Parliament. With me today is Mike Sheridan, our chief operating officer. On behalf of Canada Health Infoway, I want to thank you for the opportunity to contribute to your study on post-market drug surveillance.

Since we don't have too many opportunities to appear before you, I'd really like to take the liberty of telling you about some of the important work we're engaged in, work that I believe will transform the delivery of health care in Canada.

Created in 2001 by Canada's first ministers, Infoway's mandate is to accelerate the development and adoption of electronic health records--or EHRs, as we call them--across the country. Infoway is an independent not-for-profit corporation whose members are Canada's 14 deputy ministers of health. We are funded by the federal government and operate as a strategic investor with all 13 provinces and territories, jointly investing in core systems across the country. I will explain in a moment what these core systems are.

First I will say that the collaboration has been unique and a remarkable success story. By working together, provinces and territories are sharing best-practice designs and systems, which have dramatically reduced the cost, the time, and the risk.

Having read some of the testimonies before the committee, I note that previous presenters suggested our emerging EHR infrastructure could be part of the solution of effective post-market surveillance of drugs. Infoway certainly agrees that the potential exists, but as I will outline, there are some challenges.

The good news is that while our solutions have not been specifically designed for surveillance, they are already contributing to reduce adverse drug events.

I have a simple example, if you will allow me to share it with you. In Ottawa one evening a senior enters an emergency ward, confused and disoriented. Physicians need to stabilize his condition with drugs, but first they need to know which other medications the senior may be taking. Fortunately, today every emergency room in Ontario now has access to the drug profile viewer, which enables authorized physicians to view the medication profile of every recipient on the Ontario drug benefit plan, thus preventing dangerous interactions and adverse drug events before prescribing or dispensing.

I would like to explain some of the core systems that Infoway and its partners are introducing, but first, as a precursor to everything else I say, we recognized from the start that the success would depend on privacy and security being fundamental to all the plans we develop, to all the technology we design, and to all the systems implemented by the provinces and territories.

Our chief privacy strategist works closely with her counterparts at the federal level and with the territories and provinces. We have a team of senior engineers dedicated to designing privacy and security best practice into our core architectures, which are the basis of specifications that jurisdictions will use with their vendors.

Each project must carry out a privacy impact assessment that examines the solutions against the privacy requirements of that applicable province or territory.

To support the data needs of public health officials, researchers, and policy-makers, the systems are being designed to accommodate the identification. This would allow data to be accessed and studied anonymously, providing a wealth of health indicators.

So where are we today? Each province and territory has established a detailed three- to five-year road map to build the foundation of the electronic systems they need. Almost 260 projects are under way, representing an investment by Infoway of approximately $1.5 billion, or 95% of our total funding.

Jurisdictions' contributions, I should add, for development, deployment, adoption, and ongoing maintenance often represent multiples of this amount.

The bedrock of electronic health records are five complementary clinical information programs, or the core programs, which we jointly invest in. Each program on its own is delivering important benefits to Canadians and our health care system. Together, they capture a patient's comprehensive medical history. This is where we ultimately need to be, where all clinicians have all the right information at the right time to deliver safe, efficient care.

The first program is our registries programs, basically a sophisticated electronic directory that unambiguously identifies patients, health care providers, and in some jurisdictions health institutions.

Our next program is the diagnostic imaging program, focused on digital storage, retrieval, and sharing of a patient's X-rays, ultrasounds, MRIs, and CT scans. Going digital eliminates the cost and the inconvenience of handling film. It allows radiologists in urban centres to service remote or under-serviced locations. It has increased diagnostic speed and integrity, while improving the productivity of our radiologists--and they are pretty scarce.

Our next program is the lab information system, which allows clinicians to electronically capture and view lab results and reports from hospitals and community and public health laboratories. This reduces the time for diagnosis and eliminates duplicate tests.

The drug information system represents Infoway's fourth clinical program. Drug systems allow prescriptions to be sent, viewed, dispensed, and confirmed electronically. When they are fully implemented, they will automatically flag to the prescribing physician and dispensing pharmacist the potential dangerous drug-to-drug interactions and allergic reactions associated with a particular drug.

When Infoway began its drug investment program, very few provinces and territories had plans for a system that would provide all these capabilities. Over the last year, however, the strategies of collaboration, development, and shared cost have spurred most jurisdictions to undertake drug information systems that will cover all drugs for all people, which is a very important development.

Our last, and in many ways most important, program is the interoperable EHR, or the glue that hangs some of these other programs together. It consolidates an individual's health information from a variety of sources, including the ones I've outlined, into a single secure and integrated health record. Depending upon funding considerations and jurisdictional readiness by 2010, we're very hopeful that the interoperable EHR will be available for 50% of Canadians.

Let me close with specific issues on post-market surveillance. As you well know, the complexity of drug monitoring is exacerbated by the explosion of new products and by the aging population living with multiple chronic diseases and taking several different drugs. Drug trials typically target a limited population over relatively short durations. Often they lack real-world exposure. Analyzing de-identified data sources from EHRs that contain prescription information, examination findings, lab reports, diagnostic test results, and other patient outcome information at the population level would allow benefits and risks to be more rapidly and effectively assessed. Subject to privacy considerations, technically de-identified data could be loaded into an aggregated database in a format that allows analysis using various reporting tools.

In the future, it may be possible early in the drug life cycle, or at any point, to track efficacy and patient safety across a wide population.

Now comes the bad news, because having said this, I must caution the committee that, first, our current plans and funding do not include the tools or the required analysis systems to do post-market surveillance studies. Second and more immediate, although completion of our current goal in 2010 represents a significant milestone, it represents less than half of the EHR solution.

To finish what we've started takes commitment, and unfortunately it takes money. Two recent studies estimate that a total of EHRs for all Canadians in all settings would be about $350 a Canadian, or about $10 billion spread over 10 years. The promising news is that these same studies confirmed that once fully implemented, electronic health records will deliver savings estimated at between $6 billion and $7 billion each year, money that can more productively be reinvested in other priorities, whether they be health care, education, innovation, or infrastructure.

In conclusion, Canada is implementing a powerful health information platform whose driving force has been better health care for individual Canadians. Once in place, it may present opportunities for building secondary applications such as post-market surveillance for drugs.

Mr. Chair, that concludes my opening remarks. I would be delighted to answer your questions.

[The Vice-Chair, Liberal, Lui Temelkovski]

Thank you very much, Mr. Alvarez.

We will now start the seven-minute question and answer period with Dr. Bennett.

[Carolyn Bennett, Liberal, St. Paul's, ON]

Thank you all very much.

My concern has always been that the technology is there to actually help us do this right away. If we look to the veterans administration in the United States, from the worst care in the country to the best care in the country, most of that was transformed because of their insistence on an electronic health record in a hurry.

I have huge concerns that even though the technology is prepared and able to encrypt the data, the examples that keep being used around privacy are ones where the data was not encrypted, and that privacy bogeyman keeps getting in the way of our getting on with what we need to do.

If we remind ourselves of Judith Maxwell's very important work during the Romanow commission on what patients and Canadians think about privacy, where they are more than willing to let their family doctor let the consultant know what's going on, we have been hindered sometimes by various health professions not wanting their prescribing practices tracked, as opposed to the patients, who actually do want to know whether their information being shared without the identifiers on it could make a safer system. I think most Canadians would want to be part of that.

If we go forward—and we heard in this committee last week that it's the written prescription that means the diagnosis can't be put on it, because it's a scrappy piece of paper that can fly anywhere and have the diagnosis and the prescription on it—I would want to know, first from the Privacy Commissioner but also from Infoway, can we and could we, if we had the resources, go immediately to electronic prescribing that included the diagnosis, which would actually help us with off-label prescribing and would help us with everything from recalling a drug to all the things that we are worried about in this real-world safety, of what we're talking about in post-market surveillance?

[Jennifer Stoddart]

Honourable member, that is a very specialized question about the role of diagnostic information. I would say that it probably could be done if you took the appropriate privacy precautions. The issue would be, is this information personally identified? If it's not personally identified, if it's de-identified, is it identifiable? How easily would it be identifiable, and who is it being shared with? What are the security considerations around it: the encryption, the safety procedures, the security procedures, and so on?

Properly done, this might be possible. Not properly done—and that's why we are working closely with Canada Health Infoway and the provincial commissioners—this could be very injurious to Canadians if their various personal diagnoses are found in trash cans in alleys, as is now happening with health information.

So you understand that it's a severely qualified answer.

[Carolyn Bennett, Liberal, St. Paul's, ON]

Yes.

Given the investments that we are prepared to make with Infoway and with the provinces and territories around encryption and around using the technology possible--two key systems, whatever it is--with due respect, I'm concerned that this is now being discussed between privacy commissioners and technical advisers of Infoway, and that the citizens of Canada are not actually being involved in a meaningful way in a citizen engagement process by which they could determine what risk they are prepared or not prepared to have in order to get the best care for themselves possible, as well as the safest possible system, including post-market surveillance.

Is there a citizen engagement process to deal with this issue of privacy and health records?

[Jennifer Stoddart]

We do polling on an annual basis. That is the extent of our direct consultation on this area. Perhaps Canada Health Infoway and the provincial commissioners, who are not here today—

[Carolyn Bennett, Liberal, St. Paul's, ON]

With due respect, I don't think polling has any relevance on this, because it just depends on what kind of question is asked. A proper process involving deliberative dialogue, in terms of the risks and the positives, is the only way that something as sensitive as this can be determined. I would like to know whether Infoway has the resources.

How do privacy commissioners determine what is the value system of Canadians without talking to Canadians in a deliberative fashion?

[Jennifer Stoddart]

On that answer, honourable member, we rely on the laws that I have to administer, and their interpretations. Regarding the complaints that come before us, I remind you that we have many complaints. We've intervened, and I just gave the example of the very important complaint dealing with the release of a field of information--which was the province--in the CADRIS database, and our role in intervening on that. So we do this indirectly. For the moment, my office is not equipped to run general personal consultations on something that, with due respect to the question you asked me, is a very specific question about diagnostic information.

What we know from documents, from being active in this field, from talking to our partners, is that Canadians value both. They value their health. They value their health system. They also value their privacy. They're willing to make arrangements--even, one might say, compromises--between the two, as we all do in balancing our privacy values with the other things we may want to achieve in this society.

[Carolyn Bennett, Liberal, St. Paul's, ON]

I can remember that one of the realities of being a family doctor was that I happened to be up all night delivering a baby when a drug was recalled, so I didn't see the news that night. The patient who came in first the next morning had seen on the news that a drug that they were on had been recalled. At that point, I had no way of sorting out in my office which other patients were also on that drug, other than by memory.

We know that it would be possible with a proper system to involve the patients. And certainly, the heart-rending testimony of one of the witnesses whose daughter was still on the drug months after it had been recalled....

So in terms of patients and Canadians being polled on this information, they need to hear the stories that would inform perhaps a broader understanding of what privacy really means.

Maybe, Dick, you can tell us how far you are on being able to get all the doctors hooked up so that we can recall a product in an effective way. Or do you think that will be done through the pharmacies?

[The Vice-Chair, Liberal, Lui Temelkovski]

In 30 seconds or less.

[Richard Alvarez]

You've asked a lot of questions. Let me just stop and say, look, we're trying to run a 21st century health care system with 19th century paper. And there is absolutely no security in the paper world. Papers are used on movie sets and they fly all over Toronto--actual medical records. We've had those types of examples.

So yes, we have done discussions and polling with Canadians. Ninety percent of Canadians want the electronic health record as long as their privacy considerations are looked after, and they look to the people they put in place to make sure those are looked after. When they're told there be will an audit trail, that if there's unauthorized access they will be informed about it, the comfort levels certainly go a lot higher.

So from a privacy perspective, we are putting in a whole host of things--user identities, user authentication, access control, and a whole bunch of things. But the way the systems have been designed, you have a client registry that has demographic information; it's not a lot of use to most people. With the labs, the drugs, etc., not only is the data encrypted in there, but it has a code that has to get hooked up to the client registry before you can find out whose information it is. So even if you hack into those systems, you're not going to be able to find out whose information it is.

[The Vice-Chair, Liberal, Lui Temelkovski]

Thank you, Mr. Alvarez.

We've gone a minute or so over.

Monsieur Malo.

[Luc Malo, Bloc, Verchères—Les Patriotes, QC]

Thank you, Mr. Chairman.

Thank you for joining us today.

Mr. Alvarez, I would like to come back to something you said in your conclusion. You quoted some figures that are sure to please the Auditor General, and perhaps she will be tempted, during the course of her study on electronic health care, to take a closer look at them. However, I'd like you to elaborate further on your comments.

You told us that while it would cost $1 billion a year over ten years to implement an effective system, the resulting savings would be in the order of $6 to $7 billion a year. In addition, this would put an end to adverse events in relation to prescription drugs and post-market surveillance.

I would simply like to know how you came up with these figures. If they are conclusive, how is it that everyone who hears them is not proclaiming this as the solution for the future? For an outlay of $1 billion a year over 10 years, we can save $7 billion a year. It seems to me that if these figures were accurate, we would already have opted for this course of action.

Could you elaborate further on this scenario.

[Richard Alvarez]

Veuillez m'excuser. I'm going to answer in English, but my colleague over here might want to add something.

First, these numbers haven't been pulled out of the air. The numbers have basically been derived from two major studies, one by Booz Allen, one by McKinsey & Company. Also, we have similar sorts of numbers when studies have been taken in the U.S., certainly when projects like this have been rolled out in the NHS system in England, and certainly in the Scandinavian countries, etc., and in Australia when they've done the numbers.

The simple fact is this. For years we haven't invested in these systems, because it's very difficult at times to sell these infrastructure systems to the public. It's much easier to sell another doctor, another nurse, another MRI machine, another piece of equipment. If you're really going to transform the system, then you have to have some evidence-based medicine in which to work, and this is the way it brings you evidence-based medicine, by having these technologies.

Look at the financial industry. It would not be able to perform today without the kind of computerization it put in. Let me tell you, that in itself took 20 years and a lot of money. When they transformed that industry, they were spending anywhere up to 12% of all their revenues. Today they're spending probably about 6%, because the systems, the ATMs, are there.

In Canada we're spending anywhere between 1.5% and 2% on information technology, and that's right across the country. We can't make these kinds of changes unless we get it up to about 4%, but it's a very tough sell to be able to do that.

In terms of the benefits, I don't think treasury's going to take out that $6 billion and $7 billion. I can tell you that with the tsunami of our aging population and chronic disease, they really are going to be in a position where we can expand the capacity, improve the access, if we put these systems in place.

Mike.

[Mike Sheridan, Chief Operating Officer, Canada Health Infoway]

As Mr. Alvarez said, EHR systems have not been implemented everywhere in Canada. Nevertheless, to give you some idea of the benefits to be had, I would just mention diagnostic imaging. According to the follow-up studies that have been, the productivity of radiologists has increased by 20%.

These studies weigh other possible program benefits, and the potential for using funds. When we look at everything, at amalgamating drug systems, diagnostic imaging systems and laboratory systems on a Canada-wide basis, the potential for substantial savings is clear.

[Luc Malo, Bloc, Verchères—Les Patriotes, QC]

Would you care to comment on that statement, Madam Auditor General? Can we in fact do this kind of cost-benefit analysis using the information currently available to us?

[Sheila Fraser]

Obviously, I cannot comment on this particular matter because we are just beginning our audit.

Given the scope of the project, we would expect a cost-benefit analysis to be done, because we are talking after all about several hundreds of millions of dollars.

As Mr. Alvarez mentioned, we can draw on the experiences of other countries and adapt these to the Canadian reality.

[Luc Malo, Bloc, Verchères—Les Patriotes, QC]

Will you be focusing on this area in your audit?

[Sheila Fraser]

I would assume so. Perhaps my team would not want me to commit to doing something that wasn't planned, but this is certainly the kind of information we would include, whether the organization did a study and assessed potential. However, we would not be verifying these studies or giving them any credibility. It would be more a matter of making a passing reference to them and nothing more.

[Luc Malo, Bloc, Verchères—Les Patriotes, QC]

May I ask the Privacy Commissioner one small question?

Some of the witnesses who testified before the committee talked about doing more work with the human genome in an effort to establish a causal link with respect to drug adverse events.

Could the use of the human genome in studies of this nature have privacy implications?