Episode 19      30 min 32 sec
Individual Privacy vs Public Benefit in Health (Medical) Informatics

Prof Loane Skene and Assoc Prof Jim Black discuss the debate on medical informatics with Sian Prior

Guests:

Professor Loane Skene, Chair of the Board of Undergraduate Studies, and Professor of Law at the Melbourne Law School and the Faculty of Medicine, Dentistry and Health Sciences
Assoc Prof Jim Black, Epidemiologist and Public Health Physician at the Nossal Institute for Global Health

Topic: Individual Privacy vs Public Benefit in Health (Medical) Informatics

"... there are many situations in which doctors are required by law to report things about peoples' health." - Prof Loane Skene




           



Professor Loane Skene
Professor Loane Skene

Loane Skene is a Professor of Law in the Law and Medical Faculties at the University of Melbourne and Chair of the University!|s Board of Undergraduate Studies. She came to the university in 1992 after working as a solicitor in Melbourne and England; and a policy adviser in Canada and Melbourne (10 years with the Victorian Law Reform Commission). She specialises in Health and Medical Law, is the author of two books on law and medical practice, and numerous chapters in books and articles. She has served on many federal and state advisory committees, especially in relation to the legal regulation of genetic testing. In 2005, she was Deputy Chair of the Lockhart Committee on human cloning and embryo research. In 2003, she was awarded a Centenary Medal for 'Service to Australian Society through the Exploration of Legal and Ethical Issues of Health Care'.

Assoc Prof Jim Black
Assoc Prof Jim Black

Associate Professor Jim Black is an epidemiologist and public health physician, and has recently joined the academic staff of the Nossal Institute for Global Health at the University of Melbourne. He spent ten years of his early career in Mozambique and Tanzania, as a clinician and epidemiologist, and later spent six months setting up disease surveillance and outbreak preparedness in East Timor. After completing a PhD from Monash University he spent five years as head of epidemiology in the Victorian Infectious Diseases Service. This included the development of the SynSurv pilot project in syndromic surveillance using real-time hospital emergency department data.

Credits

Host: Sian Prior
Producers: Kelvin Param, Eric van Bemmel and Sian Prior
Audio Engineer: Craig McArthur
Theme Music performed by Sergio Ercole. Mr Ercole is represented by the Musicians' Agency, Faculty of Music
Voiceover: Paul Richiardi

Series Creators: Eric van Bemmel and Kelvin Param

Melbourne University Up Close is brought to you by the Marketing and Communications Division in association with Asia Institute.

View Tags  click a tag to find other episodes associated with it.

Download file Download mp3 (29.4 MB)

Individual Privacy vs Public Benefit in Health (Medical) Informatics

VOICEOVER
Welcome to Melbourne University Up Close, a fortnightly podcast of research, personalities, and cultural offerings of the University of Melbourne, Australia. Up Close is available on the web at upclose.unimelb.edu.au That's upclose.u-n-i-m-e-l-b.edu.au.

SIAN PRIOR
Hello and welcome to Up Close, coming to you from Melbourne University, Australia. I'm Sian Prior. Today we are going to tackle the topic of 'Health Informatics' or as it is sometimes called, 'E-Health' or 'Medical Informatics'. These are all terms used to describe the use of information technology to collect, anlayse, store and disseminate health information. On the positive side, we are told health informatics has the potential to revolutionise health care. For example if there was a major outbreak of a dangerous infectious disease in your community, health informatics could help medical authorities to monitor its progress, disseminate information and coordinate patient treatment. E-health could also ensure that in a medical emergency, your health record is on hand wherever you are: at the GP, with a medical specialist or in an emergency ward, leading to better communication and more streamlined medical care. But there could be a high price to pay for these benefits. The ready availability of this information could mean a loss of privacy for patients. It could also lead to a patient's medical records being available to people like employers and insurance companies who might use that information to the patient's disadvantage. So, how do we find the right balance between maintaining civil liberties and potentially saving lives in the era of health informatics? With us today, are two people at the forefront of this discussion. Loane Skene is a Professor of Law in the Law and Medical faculties, here at the University of Melbourne, Australia. She has previously worked as a solicitor in Melbourne and England, and is the author of two books on law and medical practice. And numerous chapters in books and articles. And, Associate Professor Jim Black is an epidemiologist and public health physician who has recently joined the academic staff of the Nossal Institute for Global Health at the University of Melbourne, Australia. He spent ten years of his early career in Mozambique and Tanzania as a clinician and epidemiologist, and later spent six months setting up disease surveillance and outbreak preparedness in East Timor. He also spent five years as head of epidemiology in the Victorian Infectious Diseases Service. Welcome to both of you.

LOANE SKENE
Thanks, Sian.

JIM BLACK
Thank you.

SIAN PRIOR
Well Loane Skene, I wonder if we could begin, if you could give us just a brief summary of the kinds of information that are most likely to be collected, analysed, stored and disseminated in this new era of health informatics.

LOANE SKENE
Well, I think you have summed it up in your opening, Sian, in talking about information that is collected for clinical purposes to assist the health treatment of patients, and information that is collected for research. An example of the second type of information is the Western Australian Birth Defects Registry. And, in Western Australia, information is passed on by medical practitioners on a voluntary basis if a child is born with a birth defect. And that is done without the patient's consent. Because it is recognised that there is a real public interest in collecting information about everyone in the whole population. If you ask people for their consent, that may slant the sample group that you get. Once you have got that information about birth defects, it may be possible to link that data with information that has been collected for other purposes. So, you might find out more about the health of the women concerned. It might be information of an employment nature, so you might be able to find out that there is something in their employment or in their environment that is contributing to children with birth defects. You can see at once the importance of this sort of information being collected. In Western Australia, which is a very large area, with a low population and so each doctor might encounter only one of these cases in the whole time that he or she is practicing. So it is important to be able to collate these rare instances of birth defects in this case, across the whole of the state and put them together and then look for other factors that they might correlate with. There are lots of other types of information that has been collected in this way. In the United Kingdom for example, information was collected that showed an association between folate and neural tube defects. And it was collected in the same way, with information being obtained from many sources and then put together and links looked for. Another example in the UK, is the UK bio-bank. And, at the moment, this will include, when it is fully established, 500,000 people, who are contributing their genetic material, like blood samples for genotype and are also answering a lot of questions about their health and lifestyle for phenotype.

SIAN PRIOR
What might be the benefits of that information being stored?

LOANE SKENE
Yes, well again, they're hoping to find links between what people eat, how they live their lives, how they exercise, the sorts of jobs that they have, and their family history, because people will be asked about the family medical history and that will be associated with conditions like cardio-vascular conditions and cancer. So, it is an exercise to try and find out more information.

SIAN PRIOR
Jim Black, you're an epidemiologist, and a public health physician and epidemiology is one area where, presumably, health informatics has enormous potential, what sort of differences has this use of information technology meant for people working in your field?

JIM BLACK
I think it is more a matter of what hasn't been done yet. That we've got all the technology that we need to do much better studies than we've been able to in the past, but we actually haven't been able to put them together. Western Australia is a really interesting example because it is the only place in Australia, where there is that serious attempt to link all the necessary data sources. And we are talking about things from the deaths registry, through to hospital information and prescribing data from GPs making prescriptions in their practices and so on. Now, we are always interested in what is happening in the larger population. There is a kind of joke amongst epidemiologists that nobody ever gets to do one of those big cohort studies, as we call them, because by the time you are well known enough to start one and get the funding for it, you are too old to see it through to the 20 years that it takes do it, and anyone who is young enough to start one can't get the funding because they don't have the track record of that kind of study. So, there is an obvious solution: that is, start up a data base, that basically follows everybody, but it means everybody, with no opting out, and that way you can be sure that you've included the really interesting and difficult ones and that you'll get information on everybody as you go forward in time.

SIAN PRIOR
What has stood in the way, thus far, of people other than, people in Western Australia, being able to access that kind of information?

JIM BLACK
Part of it has been the technology, it is only now that the computers are fast enough and the storage is big enough and so on, that you can seriously talk about merging data sets and linking them. But part of it is that, either there is some statutory reason why that information can be collected, births and deaths must be registered or they have given that information for some other purpose !V like, so that the doctor can diagnose and treat their illness and they haven't given permission to be involved in a study. And it is quite hard to get those people who hold those data sets to allow researchers to put them together. They might be happy to give access to a de-identified portion of the data, but to say, 'no, I actually need to know who each of the people is, so that I can identify them in the other data set', it is actually very hard to get people to do that.

SIAN PRIOR
So, de-identified means you just don't actually take or keep personal details, the name, all that kind of stuff, for each individual patient.

JIM BLACK
In fact, with my epidemiologist hat on, I would really like everyone to have a barcode somewhere on their body. Readable, so that you can't cheat, and you can't lose it, and you can't accidentally be given two of them and so on. Because name and date of birth are actually quite difficult to use for those purposes of linking people. The ideal would be a single identifier !V

SIAN PRIOR
People could change their names for example.

JIM BLACK
They do. I saw a simple example in Tanzania, in a study I was doing we had a regular census going on, in which we would come back every six months and basically be asking, 'has anyone new moved in?' or 'has anyone been born? Has anyone died?' And whenever someone had died, we interviewed their next of kin to try and work out what the circumstances of their death were in an attempt to work out what the overall rates and causes of mortality were there. But we had another set of data, in which key people in those communities would be asked to keep an eye out and let us know if they heard of somebody dying. And then we tried to match the two groups. And we'd find that somebody would be at home, he'd be 'Ismail Muhammad', but if he wanted to go to school, he'd find that the most attractive school was actually a Christian one, and he'd become 'Joseph Charles' when he went to school. And the differences were so huge that we just couldn't match people.

SIAN PRIOR
So, there is a fundamental problem with the reliability of the data and therefore the usefulness of the data?

JIM BLACK
Yes. Those were particular circumstances in Tanzania, but something like it exists in Australia.

SIAN PRIOR
Professor Skene, Jim Black is probably only half-joking when he talks about giving us all barcodes, what are the ethical concerns about making freely available, in a compulsory way, information about peoples' health?

LOANE SKENE
Well, people are very worried about personal privacy. And this varies from one country to another. So, I believe in Singapore, they have he next best thing to the individual barcode that Jim has described, with everybody having an ID number, which they use for everything, and it is just accepted that this is the way you transact business and interact with other people. In America, there is a great emphasis on individual privacy. And that is a predominant value. In the UK, there is more of a balance, between individual privacy and the public good. And I think Australia is sort of between the UK and America. So, when there have been debates about us having some sort of national identifier, there have been many people who have said, 'this breaches my privacy. I'm not confident that my personal information will be properly looked after. It may get into the wrong hands, and I may suffer if I'm trying to get life insurance, for example, if the insurer knows that I have a family history !V or particularly if I have a personal incidence of genetic defect or some other medical condition.'

SIAN PRIOR
And that is a legitimate concern, isn't it?

LOANE SKENE
Absolutely, it is a legitimate concern. However, when surveys are done and people understand the significance of collecting community wide data and linking it in the way that I've described with regard to the Western Australian birth defects registry and they see the potential good, although some people will say, 'We're worried about our privacy'; other people will say, 'Well, I'm surprised that the scientists haven't been doing this sort of linking for years. This is the most basic type of research that you would expect any epidemiologist to be doing, any government to be doing.' So, I think we have to look at the concerns. I think there is a certain level of ignorance in the wider community about why this sort of valuable and should be done. But is also equally important that the technology is reliable. And that we can trust the people that are collecting and collating the information and that we can trust the coding and the other technological safeguards when the information is being collected, stored, linked and retrieved.

SIAN PRIOR
And that is a very difficult issue, isn't it? Because people read in the newspapers or find on the net, examples of information going where it wasn't originally intended to go. We are talking about information being able to be sent, in digital form, anywhere across the globe in an instant, how on earth, can people feel confident, that if they're providing information to one particular organisation it is not going to go without their consent to another organisation, that they might not want it to go to, Jim?

JIM BLACK
Maybe it is even more complex than that. Let's say you had an eminent, academic epidemiologist, who was going to Canberra for a meeting of surveillance experts in a couple of weeks and when he is there !V this is not me, by the way, not even a friend of mine or a cousin !V but, let's say while he is there, he collapses and is unconscious and it turns out he has got a problem with the rhythm of his heart, atrial fibrillation, so he goes to a hospital and there are very good doctors there ready to help him and they know that if you have just started having atrial fibrillation, then the best treatment is probably to give an electric shock to the heart and that will often switch back into a normal rhythm, but if you have had it for more than a fairly short time, then that could be quite dangerous because it might dislodge a clot that has formed on the part of the heart that is not beating properly. And then, if you have had the disease for a time, then you're probably on some medical treatment, and so obviously the worst thing you could do in that case, would be to give more of the same drug. And many people would be asking as Loan suggested, 'how come we can't get that information?' If that person's information is carefully stored in digital form in Melbourne, or if I could check the cinema starting times in Melbourne from Canberra, why can't the doctors check on the patient's records? In many of the hospitals in Melbourne, if not all, even the ECG recording will be in a digital form. So I think many people quite rightly would say, 'why are we not doing that, already?'

SIAN PRIOR
But that hasn't answered my question, has it?

JIM BLACK
No, but there is another side to the story. What if that same eminent person had been a heroin user, in his undergraduate years and had contact with the health service somewhere and there is life after heroin and you'd be surprised about how many perfectly normally, functioning members of society have a little secret like that in their background, and let's say he had a minor psychotic episode, again, during his undergraduate years !V it is not that uncommon for people to have one brief episode of psychosis and never again !V now, that may be completely irrelevant to all the future medical treatment that he has and he may not be happy for every health worker that has access to that beautiful stream of data, to be able to see all of those things as well, but how do we decide which things might one day be relevant and which things might not be relevant and how do we have the safeguards built in even if it wasn't possible to take those data anywhere else other than within the medical sphere. There are still some uncertainties.

LOANE SKENE
Jim, I think, what he's worried about is not that these health workers are going to see this information, but that somebody outside, for example, the media, if it is somebody in public life, might see this information. And to answer Sian's question about how we are going to keep this information confidential when we know about leaks of other sorts of information, you could really turn it around and say, 'when you think of the enormous amount of information that is circulating including information on the internet, it is surprising that there aren't many more breaches than there are. I think as we move forward there are going to be new methods of protection that we think about. I'm confident, that first of all, we can work out ways to protect the security of the information and secondly, if there are problems that arise later, that we'll be able to put in new methods of protecting the information. So implanted barcodes, of the kind that Jim is describing are another !V and people remember the Holocaust and that tattooed numbers on the Nazi prisoners.

SIAN PRIOR
Yes, some people have some very, very negative associations with those kinds of methods of identification.

LOANE SKENE
That's right. And that is why it is so important for people in public health to be educating the broader community on why it is a good idea.

SIAN PRIOR
I know I'm harping on this point, but it seems to me that relationship of trust between a doctor and a patient, that relationship of confidentiality that you can go to your doctor and that you can tell them things about yourself, about your health, about whether or not you are pregnant and might be considering an abortion for example, a highly controversial issue like that !V people are used to being able to assume that that information will go no further.

LOANE SKENE
Well, if they assume this they are mistaken, because there are many situations in which doctors are required by law to report things about peoples' health. And so if people have notifiable diseases like HIV for example, if people are suspected of abusing their children, if people are being prescribed drugs of dependence and there are many other examples where doctors can't maintain complete confidentiality and they are already required to report these things. Obviously, only to the relevant government departments, but the doctor will be fined if he or she doesn't make this type of report. And when you think of the large number of reports of this kind that are being made in order to protect people from infectious diseases, to protect the partners of people who have sexually transmitted diseases, to protect children who are vulnerable and at risk of child abuse, and yet we don't see this information in the press or being misused.

SIAN PRIOR
We are already willing to make exceptions, aren't we?

LOANE SKENE
Exactly.

SIAN PRIOR
I'm Sian Prior and you're listening to Melbourne University, Up Close, where today we are speaking with Professor Loane Skene, Professor of Law in the Law and Medical faculties, here at the University of Melbourne, and Associate Professor Jim Black an epidemiologist and public health physician who works at the Nossal Institute for Global Health, here at the University of Melbourne, Australia. So, Loane Skene, you talked about the fact that the US, in particular, has a very strong concern with privacy, what impact is having on this field of health informatics and the law in the USA?

LOANE SKENE
Yeah, well, I think Sian, one good example of this is in genetic testing. If you have a genetic test in the United States, there is legislation !V Genetic Privacy and Non-Discrimination Legislation - that gives you the right to veto that information being given to any family members. In Australia and also in the UK, it is recognised that some types of information have important implications for family members. And so, if a patient is found to have a genetic condition that has serious implications for family members !V

SIAN PRIOR
What is an example?

LOANE SKENE
An example is the colon cancer, which is called 'familial adenomatous polyposis', where growths form in the colon. If they are detected they can be removed and sometimes parts of the colon can be removed and the patient won't die. If they're not detected in time, they can spread and the patient will die. So, you can assume if a patient presents with this and has a genetic test and is found to carry the genetic predisposition for this type of cancer and the patient says, 'I don't want anyone else to know about it.' Then, the patient's close blood relatives are denied the opportunity to have their own tests and there is a serious threat to their lives. Now, in Australia and the UK, in this sort of situation, the doctor would be justified !V even without the patient's consent !V in going to the relatives and telling them there is a family history, or a family incidence of this type of cancer in your family and you may choose to be tested.

SIAN PRIOR
So, they don't actually identify which member of the family has it, they just say it's in the family !V

LOANE SKENE
That's right. And in particular, they don't say the status of this member of the family. So, the fact that you are positive or negative for the mutation in relation to familial adenomatous polyposis, often called F.A.P. for obvious reasons, your status, be it positive or negative is your personal information. But the fact that there is a risk of this within your family is considered familial and now in Australia, many doctors say, 'we treat whole families', rather than individual patients.

JIM BLACK
When you are talking about familial adenomatous polyposis, I'm thinking of the people who have discovered they are not actually related to the people that they've grown up with for all of those years. That's not the informatics, that's the simple fact of being able to do the tests. And that in itself raises all sorts of difficult issues that need to be thought through before you do the test, and people should be aware that they may discover information that they didn't want to know about.

LOANE SKENE
Genetic counselors don't have to tell them that. What they can tell them is that 'you are at risk or not of this genetic condition.' And this information can be given without saying, 'the reason you are not at risk is because you are not the child or the child or the brother or whatever'.

JIM BLACK
Yes.

SIAN PRIOR
We talked earlier about concerns people might have, for example, about health information of theirs going to insurance companies or potential employers, that is, I think, a legitimate concern. And on the other hand, insurance companies have got a lot of money riding on this, this is information that would be very much in their financial interest to be able to get their hands on. How do we ensure that financial imperatives don't dominate in these instances?

LOANE SKENE
Well Sian, in Australia, we have community rating for health insurance, unlike in the United States for example where people have to take out their own health insurance, which is usually provided through employers, but in Australia you can't be charged more for health insurance because of your condition. But with life insurance and other protection of earnings through insurance, it is true that it is based on your medical condition. If information was shared in the way that we've been describing and other information was available electronically about a patient !V again it just gives a more precise indication of a person's risk. People will still get health care, they may have trouble getting life insurance. But, a contract of life insurance is one in which in legal terms is called 'uberrima fides', 'of utmost good faith'. And people are required to tell the truth to their life insurer and if they don't they can avoid the policy. So, none of the collection and disclosure of information changes any of these principles, which go back many years. All it does is provide a clearer source of information.

SIAN PRIOR
We talked about the global nature of health informatics these days, the ability for this information to go all over the world in an instant, because of the nature of the technology that is involved are national governments, or state governments the best people to be regulating this, or is it going to require some kind of international regulation or arbitration?

LOANE SKENE
Well, I think that some sort of international scheme sounds very attractive, but legally governments really only have legislative powers within their jurisdiction, obviously countries can form together and form international agreements to enact complementary legislation, usually this sort of thing happens when money is involved, as you were saying earlier, but there is no reason why that they shouldn't do this given the will. I think we are going to need controls on the transfer of information, particularly with so much information being available on the internet, for example, which is relatively uncontrolled because it has been so difficult technically to have laws with regard to information on the internet.

JIM BLACK
One of the policies perhaps is to not put things anywhere that they can be connected to the Internet. Here in Melbourne, we've got something called the MMIM, the Molecular Medicine Informatics Model, and the people involved in that are exploring some of these issues very carefully and very effectively. For example, if you've got information about notifiable diseases, in which there is a legal requirement that it be collected and so on, none of the data bases that hold that information can be connected to the internet in any way.

SIAN PRIOR
So it is quarantined!K

JIM BLACK
It is quarantined. And the people whose job requires them to look at that database, don't automatically have any access to any of the other databases. Even though they might be interested at times to linking them together. So what the MMIM model does and it is a similar process to what they are doing in Western Australia is to say, 'we keep each of those potentially dangerous data bases quarantined, and not actually connected directly to anyone else and then we extract from them the appropriate parts with identifiers in them that allow us to link them to the appropriately extracted parts of the other data bases, and then just before making it available to other people, we trim off the identifiers.' So, we say, 'okay, this information about the Jim Black: deaths registry, and this information about drugs prescribed relates to the same person, but we'll now give them a completely new identifier that only applies to the new truncated data set.' Truly de-identified, not potentially re-identifiable information. And that is probably the technical solution that will actually work. But what it means is that you can actually say, 'we want to study people who responded well to this drug for this type of cancer, with this kind of pathological diagnosis, so we can pull out a sub-set of those people from the data sets without compromising the privacy of anybody involved.'

SIAN PRIOR
And that sounds like a very good example of what Professor Skene was talking about before. About being confident that technological solutions will be found, to the problems currently presented by technological developments. Thanks very much to both you for joining us today.

LOANE SKENE
Thanks Sian.

JIM BLACK
It's been a pleasure.

SIAN PRIOR
I'm Sian Prior and my guests today in Melbourne University Up Close, have been Prof.Loane Skene, Professor of Law in the Law and Medical faculties, here at the University of Melbourne, and Assoc Prof Jim Black an epidemiologist and public health physician who works at the Nossal Institute for Global Health, here at the University of Melbourne, Australia.

Melbourne University Up Close is brought to you by the Marketing and Communications Division in association with Asia Institute of the University of Melbourne, Australia. Relevant links, a full transcript and more information on this episode can be found on our website, at upclose.unimelb.edu.au
We also invite you to leave your comments or feedback on this or any episode of Up Close. Simply click in the add new comment link at the bottom of the episode page. This program was produced by Kelvin Param, Eric Van Bemmel, and myself, Sian Prior. Audio recording is by Craig McArthur and the theme music is performed by Sergio Ercole. Melbourne University Up Close is created by Eric Van Bemel and Kelvin Param. Until next time thanks for joining us. Goodbye.

VOICEOVER
You've been listening to Melbourne University Up Close, a fortnightly podcast of research, personalities and cultural offerings of the University of Melbourne, Australia. Up Close is available on the web at upclose.unimelb.edu.au, that's upclose.u-n-i-m-e-l-b.edu.au. Copyright 2007 University of Melbourne.


show transcript | print transcript | download pdf