Published in The Medical Post
11 January 2000
DON’T CLOSE THE SHUTTERS
by Brian Lee Crowley, Ph.D., and David Zitner, M.D.
IMAGINE A CANADA IN which census information was a secret, or in which Statistics Canada couldn’t release information about the economy’s performance unless every single business and worker agreed to allow data about their work be used for that purpose. This would be a country drowning in a sea of ignorance about itself, unable to think coherently about the past or the future and unable to develop strategies to improve health.
Yet this is exactly the kind of Canada that will exist in the health care world if Bill C-6 on electronic commerce now before Parliament were to be adopted.The effect of this bill on health care would be that individual health information could not be used for any purpose without express consent for each use.Use would be forbidden even when appropriate codes were in place to prevent unauthorized disclosure of individual information. Before we get stampeded into ill-considered and heavy-handed rules about the use of medical statistics, we need to consider why such information is in the public interest, and how much less draconian solutions exist to protect the privacy of patients.
Knowledge is power in the medical world just like any other. Knowledge about how prone certain population groups are to specific kinds of illness gives us the power to take preventive measures. Knowledge about what medications are effective in combatting disease and which are not gives us the power to treat illnesses more effectively and often more cheaply. Knowledge about the different incidence of diseases in different parts of the country gives us the power to tailor public health measures to suit the traits of each community. Such knowledge allowed us to learn, for example, that smoking is associated with lung cancer and heart disease, that Ashkenazic Jews are at risk of Tay Sachs disease and that Blacks are at risk for sickle cell anemia. This is vital information that has improved quality of life for thousands, perhaps millions, of people.
OUR RECENT PAPER “Operating in the Dark” showed that today Canadians are missing vital information about access to care and the relationship between health care activities and results. This information is essential to understand our health care system and to properly manage it.
But medical information is also highly sensitive. No one wants to think when they go to the doctor that information about their personal health could become known to others. That’s why we have to balance the need for privacy of individuals’ health information against the need of the community to understand better the overall state of health of Canadians and what measures improve it. However, it is essential and technically possible to implement rules which protect the privacy of an individual while supporting the use of that individual’s information for the welfare of all.
One way of protecting patient confidentiality is to follow some, but not all of the advice of Dr. Winston Dykeman and the College of Family Physicians of Canada. In testimony in November before the Senate Standing Committee studying Bill C-6 Dr. Dykeman suggested that patients needed to give their express consent in each individual case where information about their health record might be collected and used for purposes such as evaluating the effects of pharmaceuticals or to understand population health characteristics. Thus, the patient could be secure knowing that no researcher could include their health information in a study without their consent.
On the other hand, patients might be horrified to know that such a rule might prevent important medical advances, yet is completely unnecessary to protect their personal privacy. Without the ability to aggregate information from many patients it would be difficult to learn that a particular treatment is possibly harmful. Thus, we learned that thalidomide was harmful because reviews of charts of patients with birth defects showed the mothers had taken thalidomide.Under Bill C-6 such reviews might not be possible.
WE BELIEVE THAT it is ethically appropriate and valuable to aggregate health information about many patients, providing the individual cannot be subsequently identified without consent and that appropriate penalties and remedies are in place for unauthorized disclosure.
The problem with solutions which prevent access to aggregate information is that it becomes hugely costly to a society that wants to understand its health problems better. Knowledge is costly to acquire at the best of times. Just imagine the logistical problems in collecting a consistent and properly presented set of data from all the doctors and hospitals in the country on, say, the incidence of breast cancer, or the efficacy of surgery versus drugs on breast cancer. Imagine having to contact all patients when a population health researcher suspected that a new drug caused harm 7 years after it had been introduced. The possibility of reducing death and illness through superior knowledge would be severely reduced.
Now add to those ordinary obstacles to acquiring precious medical information the obstacle that no one’s health information, even if the person’s identity was not disclosed, could be used in any particular instance, without their explicit written consent.
WE AGREE FULLY with Dr. Dykeman when he recommends that an audit trail be introduced whenever a person’s personal and identifiable information is reviewed. People should be able to learn who had access to their individual information. We believe however, that once the information is stripped of personal identifiers that it should be available to be used for the benefit of all. Ironically, a requirement to inform individuals of each use means that a record could never be stripped of personal identifying information because the population health researchers would have to know the identity in order to obtain consent.
The alternative, of course, is precisely the solution that we have adopted for other information gathering efforts like Statistics Canada and the census. We recognize the value of this information by making it literally illegal not to supply it. It is just too useful to Canadians, collectively and individually, for us to be able to have a reliable and complete picture of Canadian society and many of its evolving characteristics. The cost of not collecting the information is simply far greater than the potential cost of loss of privacy that supplying it involves.
That does not mean that we can be indifferent to the privacy issues involved. On the contrary, both the census administrators and Statistics Canada are bound by law not to reveal information that would permit their data to be used to identify specific individuals. The safeguards are stringent, and rightfully so.
Dr Dykeman is absolutely correct when he observes,
In a world of electronic data and speed-of-light communications, privacy is fast becoming one of the most important issues facing our society. And of all the information that is being collected, there is probably nothing more sensitive and confidential than our own health data. In democratic and free society, we need to be assured that this information is appropriately protected in law.
But the key word here is “appropriately”. Naturally, there is always room to improve the rules safeguarding privacy, so as to make the anonymous nature of any health information collected as airtight as possible. But that is where we should be devoting our efforts to protect the privacy of Canadians, so that such protection does not needlessly deprive us of knowledge that can save lives and reduce human suffering.
Brian Lee Crowley, Ph.D., is the President of the Atlantic Institute for Market Studies
David Zitner, M.D., is Director of Medical Informatics, Dalhousie University