January 2003
The Tri Council Policy
Statement: Ethical Conduct for Research Involving Humans is the supreme
research ethics document in Canada. Generally speaking, it is a well-intentioned,
sincere, sensitive, thoughtful reflection on the problems and concerns
regarding the ethical treatment of human subjects2 in research.
But its birth was painful and its labor long. The first draft shocked
the research community into action and individuals, professional
and academic associations, and university administrators responded with
an unprecedented number of critical comments and suggestions for revisions.
Indeed, so activated was the community of researchers that the second draft
was forced to further revision. The final document, published in July,
1997, was one that most researchers thought they could live with.
However, what many of us failed to realize was that the document was not
just a treatise on good, ethical practice but was also a potential blueprint
for increased control over the research we could do and how we could do
it.
The 1997 Statement replaced
the separate ethical guidelines that had been used previously by each of
our three major, federal, granting agencies. It was not obvious why
it was felt that: a) medical research, b) science and engineering research,
and c) social science and humanities research needed to be brought under
one umbrella of ethical scrutiny. But a consideration of the reasons
for the integration of the ethical review processes of the granting agencies
illustrates the abstract and, sometimes, incoherent reasoning that is so
evident throughout the document.
The stated reasons for the
integration included the beliefs that fundamental ethical issues and principles
transcend disciplines, and that a harmonizing of Research Ethics Boards
(REBs, called IRBs in the US) across disciplines would be bureaucratically
and educationally more effective. Oddly, the increased protection
of human subjects was not mentioned as one of the reasons! Moreover,
no evidence was presented that showed the old way was problematic in the
first place. Indeed, true to the natural reluctance of Canadians to offend
anyone, the statements on the need for harmonization and adherence to common
principles were followed by a recognition that “The effective working of
ethics review across the range of disciplines conducting
research involving human subjects requires a reasonable
flexibility in the implementation of common principles. The Policy therefore
seeks to avoid imposing one disciplinary perspective on others.”
Thus, although it is not
clear that the Tri-Council Statement has increased public safety,
it is evident that Canadian researchers are under more scrutiny and control
than ever before, because a bigger bureaucracy, with more rules, has been
created. We don’t want to suggest that increased control per se was
the driving motive of the Tri-Council, but we do want to suggest that that
is one of the real outcomes of their new ethics statement. And therein
lies a potential threat to free inquiry. We will describe four examples.
1. First, many
REBs are asked to judge the scientific validity of research in order to
consider whether the benefits expected outweigh the harm. In the
new ethics forms developed at our university this past summer, investigators
are asked to provide details on the scientific justification of the research,
the strengths and weaknesses of the design, the analysis of the data, the
justification of the sample size, and the sample size power calculation.
This is the case even for research that has been peer reviewed and funded
by one of the Tri-Councils!
Giving members of an ethics
review board the authority to second guess the expert opinion of the grant
review panel is not only absurd but opens the door to abuse. A study
by Ceci, Peters, & Plotkin (1985) many years ago demonstrated that
the decisions of ethics committees are influenced by the purpose
of the research. In their study of actual IRBs, they showed
that judgments of the ethical acceptability of the procedures of a study
varied with the expected outcome of the study, even though the method section
was held constant across conditions. Ethics review boards should
concern themselves with the protection of subjects only, i.e., they should
ask the question, ‘Does the research violate any of the subject’s rights?’
2. Second, we
are, of course, required to provide informed consent regarding the tasks
the subjects are expected to complete, anonymity guarantees, and the like.
No argument. But now we are also directed to ask our subjects
at the end of the experiment, after they have been fully debriefed,
whether they want their data used by the researcher. Let us be clear.
We are not talking about permission to reveal personal information or to
identify which subject provided which data points. We are asked to
obtain permission from the subject to use his or her data in the data analysis.
Thus experiments that study hypotheses or theories or applications that
the subjects may object to may be put in jeopardy by selective withdrawals
from different experimental conditions. In extreme cases, this would
be tantamount to giving subjects a veto over which research could be carried
out, even though there were no ethical problems of harm associated with
any of the procedures. Should investigators be obligated to match
any potential or hypothetical political implication of their research with
the political inclinations of their subjects?
3. Our third
example is a good illustration of Lou Penner’s (2002) warning about the
local REBs’ idiosyncratic interpretations of the rules. Although
the Tri-Council Statement is silent on the advertising of payment
for research participation, our local non-medical REB, for many years,
has refused to allow investigators to advertise how much money subjects
will be paid. Participants were allowed to call to find out, and
payment information is part of the informed consent form, but the amount
must not be placed in the ad. The REB’s rationale is that money
is coercion, and coercion is bad.
As a member of the local
REB, and wanting evidence to argue with the REB, Dick Sorrentino
wrote to members of the SPSP list to canvass their opinions and university’s
policies on this issue. All of the 26 respondents who replied were
in favor of advertising the amount paid to participants, and many were
quite emphatic about it. All disagreed that money is, by definition,
coercion. Only one person said that their university’s IRB looked
upon money as coercion, but even so they were still allowed to advertise
the amount.
Most of the arguments centered
on informed consent. That is, participants should be fully informed
about the payment amount and should not have to take extra steps to find
out. For example, one respondent wrote, “I believe that most people
would like to know how much they will be paid before they call for further
information. As researchers we have an ethical obligation to avoid
wasting people’s time, and flyers with insufficient information have the
potential to waste the time of everyone who would consider the compensation
insufficient.”
Many respondents disagreed
that money is coercion. For example, one person said, “this seems
utterly absurd. By these standards, if I pay a college student as
a babysitter I’m coercing her into babysitting. It is just fair payment
for services rendered. Coercion refers to the use of excessive compensation
intended to entice people to do something they would not otherwise be willing
to do (out of fear, moral objection, or essentially any other reason beyond
‘it’s not worth my time’).”
4. As a final example,
two of our colleagues in the sociology department were engaged in a health
research project in which one of their goals was to assess the adequacy
of the current method for calculating life expectancy for native Canadians.
The research was funded by the Social Sciences and Humanities Research
Council (one of the tri-councils) and one of the partners in the research
was the federal Department of Indian Affairs. The investigators’
idea was to match death statistics with persons listed on the Indian Register.
These data are kept by another federal agency, Statistics Canada, which
initially was also interested in the accuracy of life expectancy figures.
The research got hung up
because Statistics Canada felt that it would be unethical to check whether
newly dead people were listed on the Indian Register, without obtaining
permission. Permission! From whom? Unless the silence
of the dead is taken as agreement, there is a problem. Apparently,
it was not possible to budge Statistics Canada on this point.
As the “negotiations” between
the researchers and Statistics Canada unfolded, the researchers began to
realize that the issue was hypersensitivity on the part of Statistics Canada
to doing research with native populations. One need only read the
Tri-Council Statement section on research with aboriginal communities to
realize the political sensitivity of such investigations. The researchers
ultimately decided against conducting this aspect of their research.
Unfortunately the situation
may worsen in the future. The Tri-Council has now decided that the
implementation of the 1997 Statement must be systematically policed and
standardized across all institutions where research is conducted.
Working behind the scenes without the full knowledge of the research community,
various federal agencies, vying with each other for control of the research
ethics industry, are making plans for accrediting the REBs.
According to The Society
for Academic Freedom and Scholarship which has examined the available documents
the current plan:
As Yogi Berra said, “It’s deja
vu all over again.” Whether Canadian researchers will mobilize
once more to protect their interests and be successful in limiting the
‘damage’ is unknown at this time.
Clive Seligman & Richard
Sorrentino are SAFS members
This article was reprinted
from Dialogue: The official Newsletter of the Society for
Personality and Social Psychology, Fall 2002, Vol. 17, No.2, pg.
22-24.
Help us maintain freedom in teaching, research and scholarship by joining SAFS or making a donation.
