|
Chapter Six: Displays of (SINS) STRUCTURES, INSTITUTIONALIZATIONS,
NATURALIZATIONS, SIMULATIONS) in Institutional Interpretation of Federal
Texts
“Above all, I am talking
about [the] systematic tendency to a kind of behavioral reductionism, consistently
translating matters that have to do with signification, meaning, language,
and symbolization into crude behavioral indicators, often justified in the
name of a spurious ‘scientism’” Stuart Hall, Ideology and
Communication Theory, 1989, p. 43.
Cultural reproductions (Habermas, 1984) of worldviews, ideas, knowledge,
and beliefs contributing to the creation, evolution, and operation of the
IRB system were shown in the last chapter. Turning now to the second part
of Habermas’s (1984) model of reproduction, i.e., social
integration, this idea will be utilized to explore the ways federal
regulations are reproduced in institutional handbooks, on websites and
other postings, and in the press. Social integration involves the
reproduction of norms and other patterns of social membership (Habermas,
1984).
For more than a quarter century (since the enactment of the National
Research Act in 1974), institutional IRBs have been playing a central part
in the (presumed, and overwhelmingly intended) protection of human subjects
participating in research projects. And for all the regulatory efforts,
(infamous, sometimes horrendous, yet) prominent studies continue to draw
attention to inadequacies of IRBs and the regulatory system generally, i.e.,
abuses still occur, and lapses in compliance appear to be rather frequent
(ACHRE, 1995; DHHS OIG 1998b, 2000b; GAO, 1996, and others). This might
lead one to ask what is it that makes this kind of regulation worth doing?
Until the creation of a regulatory apparatus, which was not in place
until decades after various concepts were accommodated as matters of policy
(ACHRE, 1995, particularly the Wilson letters dealing with informed
consent, and written in 1947). The rules of research ethics are “more
articulated” today (ACHRE, 1995, Discussion of Part III, p. 1; see also
DHHS [2000, Jun 6] Fact Sheet, especially section regarding Civil Monetary
Penalties; Brainard, 2000, Jun 2.) Various templates for informed consent
forms are examples of this. The University of Utah’s informed consent
template is five pages long; University of Texas at Austin’s form has
expanded during the past two years, from about one and a half pages to 10
pages as of May 2002; and the University of California at Berkeley’s model
is barely one page in length. UC Berkeley’s form is titled “Low-risk Adult
Survey/Interview Form” and is more targeted to social science concerns.
(See University of Utah, University of Texas at Austin, and University of
California at Berkeley entries in bibliography for website addresses for
each of these documents.)
Not only must there be an IRB in place where federally funded studies
are conducted, but board membership criteria are also required by federal
rules. IRBs have traditionally been required to have at least one person on
the board who is non-affiliated with the university or other organization
where the IRB operates, and one member with non-scientific interests.{179}*
As mentioned, the norm in complying with this regulation has been since
inception that the same individual serves in both capacities (DHHS OIG
1998b; ACHRE, 1995), sort of a dual token minority, a “non2.”
As pointed out by Rep. Towns (U.S. House, 1998, June 11; also see next page)
and others, this situation has served to marginalize the non-affiliated,
non-scientific person, making the member a distinct minority-on-purpose. It
seems reasonable to believe this “non2” person might be hesitant
on most occasions to articulate concerns, and on the rare occasion when
s/he might, might actually say something irrelevant (or that seems
irrelevant to other, i.e., affiliated, scientific IRB members), or
be focused on an issue outside the purview of the IRB (and over time, stop
contributing or questioning altogether). This is not to imply these members
are in a general way incompetent, but they, by categorical admission, are
less apt to know what is relevant, i.e., within the purview
of the board. Therefore, having the regulation in place has produced little
if any effect. It is another cumbersome yet meaningless activity (with the
exception of obligating non2s to attend meetings). The ambiguity
in the rule provides opportunity to negate the effects of the rule
requiring non-affiliated, non-scientific representation on IRBs.
The OHRP guidebook indicates the non-affiliate should represent the
local community, should not be vulnerable to intimidation by the
professionals on the IRB, and that the non-affiliate should be “fully
utilized by” the IRB (see footnote # 179, p. 166; see also OPRR, 1993, p.
I-7). But, rather than the rule itself, it is the interpretation and
execution of the rule that is important to consider in this situation. The
interpretation and execution have not historically (since implementation of
the IRB system in 1974) operated in a way that supports the spirit of the
rule. And this widespread (institutionalized) demonstration of resistance
to the inclusion of different points of view seems counter to the most
basic goals in academic pursuits.
Rep. Towns, in a statement during U.S. House (1998, Jun 11) subcommittee
hearings, expressed his concern about the proportions of IRB membership.
“If federal law only requires one non-medical member, couldn’t a board
dilute that member’s vote simply by expanding the numbers of other members?
Shouldn’t the number of non-medical members [be] proportional to the number
of medical members?” (p. 52; see also DHHS OIG, 1998e, p. 11).
The NBAC in May 2001 recommended nonscientists make up at least 25
percent of an IRB’s membership, and that a quarter of the members should
come from outside the institution. (A draft report issued in December 2000
had required half in both instances; see Southwick & Monastersky, 2001,
Jun 1.) George J. Annas, professor of health law at Boston University
School of Public Health, questioned the potential effectiveness of the NBAC
recommendation, saying “Community representatives have no authority and
weight when they are outnumbered three-to-one by the researchers” (as quoted
in Southwick & Monastersky, 2001, Jun 1, p. A22). Community
representatives (nons, “outsiders”) have traditionally been outnumbered by
“scientific” members by four-to-one, or more. (The same could probably be
said of the ratio between qualitative and quantitative researchers; see
NBAC, 2001, which found 75 percent of current research reviewed by IRBs is
clinical/biomedical).
Informed consent templates and the way non2s have been
developed are two of the ways federal provisions and historical development
of the system are played out at the institutional level. Next to be
considered is the domain of the local IRB in the regulatory context that
includes the federal apparatus.
Local IRB Domain
IRB responsibility falls into two main categories: initial review and
continuing review of research projects (DHHS OIG 1998e). According to
Ellis, former director of the former OPRR, in his testimony before the U.S.
House (1998, Jun 11), “The local IRB at the research site is the
cornerstone of our system of protection of human subjects. No research on
human subjects may be initiated and no ongoing research may continue in the
absence of an IRB approval (45 C.F.R. § 46, see also U.S. House, 1998, Jun
11, p. 38). This is only true of federally funded research, in legal terms,
though in “reality,” nearly all research is subjected to IRB scrutiny, and
even those studies defined as “exempt” are scrutinized, as mentioned (see
GAO 2001, for example).
As research activities have spread from secret military studies and
“atomic” medicine to university teaching hospitals, now spilling into the
private sector, IRBs have been established in these new settings (DHHS OIG
1998c). Sites include state governments, managed care organizations,
private universities, state psychiatric hospitals and other agencies,
vitro-fertilization and weight-loss clinics, genetic tests developers,
physicians, dentists, psychotherapists in private practice, industrial and
corporate programs, and hybrids.{180}*
Independent, commercially driven review boards have developed (or have
become more explicitly commercial; see DHHS OIG 2000a) to review research
protocols and even conduct research, particularly for drug companies.
Institutional Institutionalizations and Ideologies
Four themes recur in the numerous and varied writings about
organizations working from the perspective of ideology critique first
offered by Marx (Alvesson & Deetz, 1996){181}*.
These are applicable not only in the present discussion of institutional
contributions to the system, but to the evolution of IRBs, as presented
earlier, and the co-opting of social scientists, presented here and in
Chapter Seven.
Alvesson and Deetz (1996, p. 199-201) explain these four themes:
In naturalization a social
formation is abstracted from the historical conflictual site of its origin
and treated as a concrete, relatively fixed entity. As such the reification
becomes the reality rather than life processes. Through obscuring the
construction process, institutional arrangements are no longer seen as
choices but as natural and self-evident. The illusion that organizations
and their processes are ‘natural’ objects and functional responses to
‘needs’ protects them from examination as produced under specific
historical conditions (which are potentially passing) and out of specific
power relations” (p. 199) … [With respect to the universalization of
managerial interests] “particular sectional interests are often
universalized and treated as if they were everyone’s interests. In
contemporary corporate practices, managerial groups are privileged in
decision-making and research. Management is ascribed a superior position in
terms of defining the interests and interest realizations of the
corporation and thereby of wide segments of the population. The interests
of the corporation are frequently equated with specific managerial
self-interests” (p. 200) …“Central to the universalization of managerial
interest is the reduction of the multiple claims of ownership to financial
ownership. [In the cases of interest here, the reduction is to formal
authority.]…The investments made by other stakeholders are minimized while
capital investment is made central” (p. 200). [Regarding the primacy of
instrumental reasoning] “Habermas (1971, 1975, 1984, 1987) has traced the
social/historical emergence of technical rationality over competing
forms of reason. Habermas described technical reasoning as instrumental,
tending to be governed by the theoretical and hypothetical and focusing on
control through the development of means—ends chains. The natural opposite
to this Habermas conceptualizes as a practical interest. Practical
reasoning focuses on the process of understanding and mutual determination
of the ends to be sought rather than control and development of means of
goal accomplishment (Apel, 1979) … “In a balanced system these two
forms of reasoning become natural complements. But, in the contemporary
social situation, the form and content of modern social science and the
social constitution of expertise align with organizational structures to
produce the domination of technical reasoning ... to the extent that
technical reasoning dominates, it lays claim to the entire concept of
rationality and alternative forms of reason appear irrational” (p. 200).
[And with respect to hegemony] “The conception of hegemony suggests the
presence of multiple dominant groups with different interests and the
presence of power and activity even in dominated groups. The integration of
these arrangements, however, favors dominant groups and the activity of
both dominant and dominated groups is best characterized as a type of
produced ‘consent.’ The hegemonic system works through pervading common
sense and becoming part of the ordinary way of seeing the world,
understanding one’s self and experiencing needs (see Angus, 1992)” (p.
201). [Alvesson and Deetz continue,] “Several studies have shown how
employees ‘strategize their own subordination,’ achieving marginal gains for
themselves through subordination but also perpetuating dominant systems
which preclude their autonomy and ability to act on their own wider
interests (see Burawoy, 1985; Deetz, 1995; Willmott, 1993) … Willmott, for
example, has explored how ‘corporate culture programmes are designed to
deny or frustrate the development of conditions in which critical
reflection is fostered’” … (1993, p. 534 of Willmott; as quoted by Alvesson
& Deetz, 1996, p. 201).
It is apparent that regulators and researchers maintain the process even
when they find it absurd, pointless, counterproductive, and even when parts
of the process do not seem right in a moral sense, i.e., they seem
unfair, unjustified, unreasonable, doable, etc. (What power the structure
has! More accurately, what power we give to the structure!!){182}*
To gain this kind of compliance, the system must be constructed, to “seem
real” at the very least. And, it seems to be a requirement, in order for
the system to operate, that the simulation “seem right” (i.e.,
reasonable, important, etc.) though sometimes compliance is gained simply
because the system appears unavoidable, pointless to resist, or that it is
self-evident, the “way we do things” (and eventually becoming “the way we
have always done things”). No questions asked (anymore). Whying isn’t
understood (or tolerated, even within oneself) anymore. (Also see Adorno’s
comment, footnote # 67, p. 60, herein.) The idea that (some of) these
systems are unavoidable appears to be the general consensus concerning, for
example, the Internal Revenue Service (IRS), frequently described as being
as inevitable as death. Similarly, the IRB system is presented as
“inevitable” in methods classes i.e., the system has been normalized
by faculty who are in turn illuminating the naturalization path (i.e.,
culting, indoctrinating) for students. The IRB process is presented as
“required” and often simultaneously described as “unclear,” “unreasonable,”
and “nonsensical.”{183}*
The development and implementation of the rules at the federal level
take on Theory Y characteristics (McGregor, 1960, also see discussion about
effects of scientific management and classical management schools on development
of IRB system, p. 216).{184}*
However, institutional IRB members appear to differ from the federal
regulators. In institutional policy handbooks, web postings, answers to
applications, etc., a somewhat different approach is apparent, one which
implies (on the part of the institutional-level regulators) a serious (and
unwarranted) decline in the trust of others to be humane and reasonable
(Theory X, in McGregor-ese), to “discharge their authority and responsibility
in an honorable way,” (see footnote # 184, below). This situation
exemplifies a universalization of managerial interests, i.e., the
existence of regulations that involve obtaining permission to proceed carry
the implications that researchers need to be told by managers/regulators
what to do and how to do it, that they need to be controlled, etc., giving
managers/regulators “a superior position in terms of defining the interests
and interest realizations” as described by Alvesson and Deetz (1996, p.
200). Deetz (1995) states, “What might be accepted as legitimate power
differences is best represented as a system of domination, because the
empirical manifestation is that of free consent, but yet structures are
reproduced that work against competitive practices and fulfillment of the
variety of interests” (p. 164).
OUIRB Application Form: Analysis
The OUIRB application form (http://research.ou.edu/Forms/index.htm,
accessed May 25, 2002), a typical example, does not accommodate qualitative
research. For example, the form requires substantial explanation beyond
information about who the participants are, in the categorical sense. This,
in theory, would be the first and in many cases, based on federal law, the
only question a researcher should (or should be asked to) address. The
specific treatment the researcher intends to utilize is the only other
relevant issue for exemption consideration. The lengthy application form is
not only counter to anything resembling “exemption,” but also to the nature
of much qualitative research, as discussed in this section.
The authors of the form write: “Your application for the approval of the
use of human subjects should consist of eleven (11) copies of three parts:
Part I, a completed application form; Part II, a description of your
research study; and Part III, subject’s informed consent form for
participation in your study” (OU Application, p. 1). Additionally, the
author(s) state(s), “You should attach supplementary information pertinent
to this study that will help the board members in their review of your
application, i.e., questionnaires, test instruments, and letters of
approval from cooperating institutions or/and organizations.{185}*
(Of course, one might question the need to “really” develop these items
until one knows they will be needed). Failure to submit these items will
only delay your review” (p. 1, OU Application). Another managerial suggestion
that resistance is futile.
This is the only application form available for human subjects research
at OU. There is no “exemption” or abbreviated version (the only difference
for the applicant is whether s/he is required to submit two copies or 11
copies of the application, as indicated below; the same form must be
completed entirely in either case).{186}*
On page two of the application form, the author(s) write(s) “If you believe
your use of human subjects would be considered exempt from review or
qualifies for expedited review as defined in Sections 4 and 12 of the
University of Oklahoma, IRB policy and procedures (website in
bibliography), [an applicant] may submit two (2) copies of this application
for initial review. If full Board review is required, [the applicant] will
be required to submit nine (9) additional copies.” This means that if the
study is “really” exempt-able, a researcher must complete the fairly
lengthy application, and wait until it is the pleasure of the OUIRB to find
out whether or not they rule the study “exempt.”{187}*
The OUIRB (and others) seem to be confusing exempt from the process
with exempt from full-board review. Exempt from the process should
(come to) mean exempt from the process. No application, no waiting, no
interference from the local IRB.
In Part I of the application, (potential, hopeful) researcher/applicants
are asked to provide the “project” time period. It is unclear if this
question is designed to solicit an answer about the duration of the data
collection portion of the proposed study, or if it is to include other
aspects of the “project.” The authors may be asking this question as a way
of determining if the study will be subject to continuing review, i.e.,
if the “project” lasts more than one year, under federal law it must be
reviewed annually (see 45 CFR §46.109[e]). This question could be posed
differently, meeting the (presumed) need of the IRB, yet not creating a problem
for researchers (in understanding the question in the first place, and in
the impossibility of anticipating the “exact” time frame for the
“project”). If applicants were asked whether or not they anticipate
the project lasting more than a year, (a simple yes/no proposition), and
provided a reminder that if the answer is “yes” their project will be
subject to, according to 45 CFR §46.109(e), continuing review and indicate
where the applicant can obtain more information. As posed, the question is
unclear, the reason for it is unclear, and, it appears, is unnecessarily
specific.
Also in Part I of the form, the authors require the applicant provide
the number of subjects. There is no reference to the number of subjects in
a study being used as criteria for any regulatory activity at the federal
level.{188}* The reason for the question is not clear, but it does
appear to be an institutional development, i.e., not linked to any
federal rule or provision. The number of subjects is not significant,
or pertinent to the regulatory function, and particularly as it applies to
exemption considerations.
In addition to being cumbersome, Part II of the application form is
nearly impossible (and unnecessary) for qualitative researchers to
complete. A description of the study is required of the applicant, which is
to include four parts: a statement of the purpose/objectives, research
protocol, confidentiality provisions, and an assessment of the
benefits/risks to subjects. In the purpose/objectives section, the authors
instruct, “Explain the overall purpose of your study and its primary
objectives, including the importance of the knowledge expected to result”
(OU Application, p. 2). To explain the overall purpose is not a
particularly problematic task for researchers, as it is often what a
researcher starts with, i.e., an idea, a purpose for looking at a
phenomenon, a curiosity about something. However the last part, i.e.,
being asked to “include the importance of the knowledge expected to result”
is problematic (see Feller quote in Brainard, 2002, Mar 29b, p. A25, re:
sometimes not knowing the relevance of research in 5, 10, 15, or 20 years).
While knowing the importance of the prospective knowledge is problematic
for any social science researcher, it is perhaps even more true for a
qualitative researcher who may have less knowledge a priori about
what the (range of) results might be. A key difference between these two
types of research is this: quantitative researchers almost always employ
hypotheses (sometimes even directional ones), and while the researcher
doesn’t know, in advance, whether or not any given hypothesis will be
supported (or even what that support or lack of it might “really” mean),
the hypothesis creates a more “knowable” set of activities for the
researcher, a more precise “direction” for the research to take, and a more
precise (and predictable) “form” for the results (see Goffman, 1971, re:
formulaic, hypothesis-driven pursuits, quoted on p. 90, herein).
Alternatively, qualitative research (i.e., nearly always designed
without formal hypotheses) involves less advance knowledge about the
direction the research activities will take or the final form for a
research report about the activities. Asking for details about what
knowledge may result from a qualitative study are not reasonable questions
given the nature of much qualitative research (observations, for example).
In the case of observations, it is not possible for the researcher to know
what s/he will actually see; it is only possible to state in a general
sense what it is the researcher would be observing.
A similarity between qualitative and quantitative research is the lack
of knowledge on the part of either about how important any findings might
be, though both groups are asked to provide such “information” on
application forms, grant proposals, etc. (In fact, it has become “natural”
to be asked to supply such information, even if it is impossible many times
to do so.) Being required to answer such questions suggests the applicant
can somehow project benefits or knowledge that future readers might derive,
hardly possible, I argue, and (totally) unnecessary for the protection of
subjects (especially when no treatment is administered).
The research protocol section of Part II requires the applicant to
“Describe the study and procedures you will use, including a step-by-step
description of the procedures you plan to use with your subjects” (OU
Application, p. 2). This entire Part II is labeled “description of the
study,” making it unclear what the authors of this sub-section, also
labeled “describe the study,” are seeking in terms of an answer. To
describe the procedures is not particularly problematic for the qualitative
researcher; in fact, describing is simple for many procedures, i.e.,
“watching people in their natural environment,” or “asking people questions
about what they are doing.” Not much more can be said about the data
gathering procedure (Wax, 1971, especially p. 6-10; Weppner, 1977; Whyte,
1987; Gray, 1979, and 1982; Garfinkel, 1967; Geertz, 1988; Agar, 1980;
Alvesson & Sköldberg, 2000; Bantz, 1983; Brainard, 2001, Mar 9; Deetz
& Kersten, 1983; Denzin & Lincoln, 2000; Fontana & Frey, 1994;
Gubrium & Silverman, 1989; Hall, 1989; Hamilton, 1998; Klockars, 1977;
Lofland & Lofland, 1995; O’Connor, 1979; Klockars, 1979; Wilkins, 1979;
Punch, 1998; Reiss, 1979; Van Maanen, 1988). Or, if interview data is to be
used, the required (yet mostly pointless) description of the procedure
might be “I’ll ask questions.” For the step-by-step description, a
researcher might include, “I’ll ask if the participant would like something
to drink. If the participant indicates s/he does wish to have something to
drink, I’ll probe about the participant’s drinking preferences, and offer
the available options (water, soda, tea, tequila, Aqua-Velva®,
hemlock). If s/he indicates s/he is not interested, I’ll suggest we get
started with the interview. I’ll then ask the first question, ” etc.
In explaining the confidentiality section, the authors of the
application form write:
Briefly describe the
procedures you will use to assure confidentiality of the data you collect
from your subjects, specifically address whether subjects will be
identifiable from raw and/or refined data, how data will be protected from
non-project personnel (e.g., stored in locked cabinets), whether the
identifiable data will be destroyed{189}*
when no longer needed, and whether project publications (theses, papers,
videotapes, etc.) will allow identification of individual subjects.” (p. 2
of OU application form)
First, shouldn’t the (adult, un-incarcerated, mentally capable) research
participants be able to decide if they care about their identities
remaining secret? What if the participants themselves have no concern about
it? And with much qualitative data, the researcher wouldn’t know the
identities of the subjects at all, unless, of course, an IRB forced the
researcher to gather informed consent forms from subjects (who can hardly
be called “participants” in strictly observational data gathering
activities). Federal rules exempt these activities from the process, if
data is recorded in such a way that participants/subjects are not
identifiable. Once again, the local institution is intruding “more
rigorously” in areas where the federal government doesn’t tread at all; see
45 CFR §46.101(b)(2){190}*.
Qualitative researchers aren’t likely to know when data “are no longer
needed” and their field notes are not likely to need “protection from
non-project personnel,”{191}*
at least not in the ways that would typically be of interest to an
IRB.{192}*
The fourth section of Part II, titled “Subject Benefit/Risk” includes
these instructions: “Describe both the potential benefits and risks to
subjects and society that may result from their participation in this
project” (p. 2 of application form). Several (unnecessary) problems are
created for (qualitative) researchers in this regard. It is not that
benefits and risks can’t be spoken about by these researchers, but this
requirement is unnecessary for qualitative researchers in the overwhelming
number of cases simply because no risk (beyond that encountered in one’s
life generally, i.e., “minimal” risk in federal terms) exists. This
makes a contrived description the best (if not the only) possibility for
the application form. Further, explaining risk/benefits (or validity and
reliability; see O’Connor, 1979; Bantz, 1983; Bloor, 1997; Denzin &
Lincoln, 2000, and others) is done in a somewhat different language by
qualitative researchers than the same description offered by quantitative
types (Denzin & Lincoln, 2000; Rorty, 1979){193}*.
And with respect to fieldwork in particular, Cassell (1982) states
“Calculating potential harms versus benefits offers little guidance to
actors under the intense pressures of the field. Few fieldworkers wish to
harm those studied; most are eager to help them. Predicting the
consequences of one’s actions, however, is even chancier in the field than
at home” (p. 155).
The IRB requirement for applicants to provide information about the
“potential benefits … to subjects and society that may result from their
participation in this project,” is problematic for all researchers,
producing forced self-aggrandizement perhaps, but little in the way of
actual, relevant information. Far ranging social benefits, or risks for
that matter, can’t be known in advance. What a researcher hopes to find
frequently has no distinct form. This is the case in qualitative and other
forms of exploration that involve looking-at exercises, rather than
looking-for pursuits, i.e., generalized explorations without any
hypotheses (much less directional ones). Further, if the IRB
supposes researchers are capable of making sophisticated predictions about
potential benefits and risks, and honestly report about them, it seems
reasonable to assume these same researchers could accurately and honestly
report the simpler issue of whether or not the study is “low risk,” “no
risk,” or “minimal risk,” (based on definitions provided by what might come
to be known as system overseers rather than direct
overseers). Researchers are at least as capable of accurately and honestly
exempting themselves as anyone else is{194}*,
including regulators. Adding to regulatory problems is the ambiguity in the
(federal) rules themselves, and considerable more ambiguity in the layers
of (institutional and individual) interpretation of each of the rules.
Part III (OU Application, p. 3-4) is nearly an exact duplicate of Part
II, i.e., both include sections in which applicants are required to
describe the study, potential risks and benefits, and confidentiality
provisions. The difference is that in Part III (the informed consent form)
the wording of the descriptions must be in “lay language,” defined as
“easily comprehensible to the person who is being asked to sign it as a
legal indication of voluntary participation in the proposed study and every
effort should be made to limit the consent form to one page including space
for the participant’s signature” (OU Application, p. 3). Yet the description
of what is to be included on the informed consent form is twice that
long.{195}* Additionally, as Cassell (1982) points out, the
definition of “subject” should be debated in many cases, in an effort to
find better rules (for unobtrusive methods for example). As mentioned,
researchers should be asking questions related to whether they should apply
at all, rather than how to go about it, or how long the application process
will take in any given instance, etc.
Following the arguments of the qualitative methodologists cited above,
the informed consent process, as required by many institutions, doesn’t
work if a researcher intends to study the natural environment. The process
doesn’t allow for the preservation of the natural environment; it corrupts
it. Only the degree to which this occurs might be argued (see Geertz, 1988,
p. 133; Capron, 1982).
The federal government’s definition of “minimal risk” states “the
probability and magnitude of harm or discomfort anticipated in the research
are not greater in and of themselves than those ordinarily encountered in
daily life or during the performance of routine physical or psychological
examinations or tests” [45 CFR §406.102 (i)]. This would seem to allow
preservation of the natural world for the numerous studies that require it,
such as observational research, most surveys, and ethnography. Therefore,
it appears it is the institutional level of interpretation where a “zone of
unreasonable interference”{196}*
exists. People who are being observed are not participants, and they
certainly aren’t patients. They aren’t subjects. They are objects in the
natural world (see Foucault, 1972, pp. 40-49, re: the formation of
objects). That is, they are objectified (see also Reiss, 1979, p. 67),
called participants, patients, subjects, etc. However, those being
observed, interviewed, and surveyed are receiving no treatment. And, most
importantly, they are at no greater risk than that encountered in “daily
life or during the performance of routine physical or psychological
examinations or tests” as specified in federal law [45 CFR §406.102 (i)].
A related observation about the OU Application: the protected classes
(pregnant women, mentally disabled, mentally retarded, fetuses, elderly,
prisoners, and children) are listed with boxes beside each to indicate yes
or no to the question of whether or not the study involves any of those
groups (p. 2). In a similar way, a question could be added about whether or
not the study involves public officials or other (prima facie)
exemptible classes. This would make the process often much shorter for
everyone, i.e., if a study involves only public officials,
instructions could indicate it is not necessary to complete the remainder
of the application. Similar checklists might be utilized to reduce the
likelihood that the entire application would be necessary, for example the
degrees of risk as defined by the federal government, a list of exempted
categories, or a checklist of general treatment types or methods, all of
which would, logically, be best assessed by the researcher. And, this
mechanization of the process appears consistent with the “standardization”
obsession of regulators in general and other universities that provide for
review of exempt studies.
As is apparent, the OU application process, even for studies that are
exempt, is cumbersome. Required but unnecessary. When we “really” think
about it, doesn’t “seem right,” or reasonable, and hardly resembles an exemption
as we would, outside this context, define one. This situation is consistent
with the view expressed by Alvesson and Deetz (1996) quoted earlier, about
naturalizations, compliance with managerial (regulator) demands, hegemonic
acceptance, etc. What is the difference, for the applicant, between an
exempted study and one that involves higher risk or protected classes?
There are differences in the procedures for the actual study, but
very few differences in the application process as it is interpreted
and imposed by local IRBs.{197}*
To summarize, it has become a self-evident “naturalization” (Alvesson
& Deetz, 1996, p. 199) that researchers must participate in the IRB
system without regard to the kind of research it is they wish to conduct. Naturalization
concerns ways a socially/historically constructed world comes to be treated
as necessary, natural, rational, and self-evident (the “N” in SINS). The
IRB application process (even for minimal risk, non-treatment studies) has
become “normal,” even when many parts of it are considered unreasonable
(and even impossible) for some kinds of studies as shown in the application
analysis offered above.
The universalization of managerial interest and suppression of
conflicting interests, as described by Alvesson and Deetz (1996), is
apparent in the way the IRB processes are “handed down” from the federal
and institutional regulators (as well as from faculty teaching methods
classes) and “passed around” by researcher among themselves. Dissent among
researchers, collectively or individually, is nearly unheard of; though
many researchers and some regulators do complain, they don’t often act.
This is evidence of the entrenchment of the system and its processes.
The “the primacy of instrumental reasoning” as described by Alvesson and
Deetz (1996), is evident in the roots of the system, i.e., it is a
system designed for clinical/medical studies, and, therefore, it is a
system that is irrelevant and ill-fitting for many methods in use today
(particularly surveys, interviews, and observational research).{198}*
One of the primary intents of regulation is standardization (and through
that a large amount of control) of decision making procedures, and as shown
in the analysis of the application form, for example, it is unreasonable to
believe, given the range of research methods, that any single form or
procedure will fit all circumstances. This “one single system for medical
and social research” mentality is in rather stark opposition, as Alvesson
and Deetz (1996) point out, to Habermas’ “practical interest” thinking
(consistent with what I have described as “liquid and local”
conditions). The more local (i.e., “real”) the decision making
process, the more valuable, useful, practical, and possible it can be for
the participants and researchers.
Finally with respect to hegemonic behavior, the IRB system and its processes
are “hegemonically” maintained by the regulators and researchers. The rules
and procedures are not questioned by many participants, who (hegemonically)
accept the idea that rules and regulations in general work to intended
effect and/or are unavoidable. Questions (of the whying sort) are rarely
asked. When questions are posed, as discussed earlier, they often focus on
minutiae, i.e., how to revise or accomplish a procedure rather than
about the need for it (see also footnote # 17, p. 15). In addition to
the bias that rules work or are needed, the application form demonstrates
instances where common sense has been pervaded, allowing the form and the
completion and submission of it, to become part of the “ordinary” world of
research, i.e., natural. Researchers are accomplices in their own
subordination by participating in an unreasonable, ill-fitting system
operated/dominated by people who often know substantially less (and often
care even less than that) about (the details of) what is “really” going on
in a given situation, method, or field of research. Similarly, researchers
aren’t demonstrating they care enough to “risk” changing the system.
The Entrenchment of Bureaucratic Ideology
These micro-managerial interpretations of the federal rules, particularly
as they pertain to social research, constitute regulatory/administrative
imperialism on the part of (institutional) IRBs, and, often, (at the
federal and institutional levels) solutions in search of problems. The IRB
system, particularly at the institutional level, carries the implicit
suggestion via explicit discourse that researchers must be carefully
monitored—a Theory X approach (McGregor, 1960; also see Taylor, 1919/1947),
i.e., the idea that researchers can’t be trusted and must be closely
and constantly supervised, spoon-fed via “templates,” and monitored via the
institution of paperwork processes (see Brainard, 2001, Sep 28, for an
alarming example). The application process is just one form of
micro-management.{199}*
The institutional-level system, i.e., the actual IRBs, as with many
other regulatory organizations,{200}*
often posture (rather than act) in the name of fairness, equity, safety,
etc., and they do so utilizing a “lowest common denominator” mentality, i.e.,
the IRBs often address problems that are first, rare; second, problems that
existing rules didn’t prevent; third, problems that rules won’t or can’t
fix (rules written that are beyond any practical scope of enforceability or
even understandability); and fourth, rules that don’t apply to the
“problem” at hand (and the enactment and enforcement of which beg the
question of a problem, or a “fixable” one, at all). Given the circumstances
of reality, specifically that the future can’t be known, that people can’t
be controlled every instant, that interpretations across people are not
standard, life can’t be standardized, etc., it would seem that a Theory Y
approach is the only realistic orientation for the regulators to assume,
and, most importantly, it is apparently an adequate approach (based on the
overwhelming lack of problems in social research).{201}*
Additional analytical frames. According to Alvesson and
Deetz (1996), in the process of naturalization a social formation is taken
from its place and time, the “site of its origin,” and treated as a
concrete, relatively static entity. So, once the IRB is established, and
rules posted, forms designed, and uploaded,{202}*
that often signals the end of questioning (no more whying, if it was ever
allowed, or if anyone allowed themselves to indulge). The simulation is
comprised of the mundane story of compliance, mostly told textualogically
time after time. “As such the reification [of various IRB processes and the
purported “needs” for them] becomes the reality rather than life processes”
(Alvesson & Deetz, 1996, p. 199; see also Baudrillard, 1983). Within
the IRB system, for example, forms required by the IRBs at various
institutions have taken priority over the substance of direct
oversight (which, even if possible, would produce an image similar to
Baudrillard’s description of second phase images, i.e., the masking
or perversion of a basic reality, 1983, p. 11). That is, an image that is
distorted because the people being observed know they are being observed, á
la the Hawthorne Effect (Mayo, 1933).
By obscuring the construction process, institutional arrangements such
as application forms and processes, who must apply to whom, definitions and
who can create and interpret them, etc., are no longer seen as matters of
choice, but as natural and self-evident (activities we “must” or at least
“should” do, and definitions, particularly inclusionary ones in this case,
that we “must” accept, etc.). This lack of scrutiny/high level of
acceptance protects (SINS) STRUCTURES, INSTITUTIONALIZATIONS,
NATURALIZATIONS, SIMULATIONS from examination and eventually even from
visibility. SINS in this case include notions that rules are generally good
as a structure that “sets up” the possibility of institutionalizing
an application process, and the subsequent acceptance of this form and
process as natural if unreasonable, and the resulting simulation
that the process is “really” doing what it was intended to do. The SINS
contribute to the notion of the process as “inevitable.” And, very
significantly, the process becomes a part of methods class curricula, i.e.,
indoctrination into the system is accomplished, and the system perpetuated.
Foucault (1980) has suggested this (zone of) invisibility is where
“real” power lives.
Regulations are debated in public and private arenas by various parties,
but these regulations are largely developed, adopted and interpreted in
private (for example, the changes in the OU Application form and other
changes, see footnote # 197, p. 185; and footnote # 202, p. 190). Because
meetings are closed at OU, input from researchers about policy issues, including
the application form, is rarely if ever solicited or given at the local,
institutional level.{203}*
For these administrative activities to be conducted in an open
environment, i.e., open to all participants who are interested (and
for researchers to require this where it is not “done this way”) would
appear to be desirable to all involved in the system. A provision, similar
to that in the federal system, to solicit and allow for written responses
to rule changes might be usefully adopted by local IRBs (or, again,
demanded by researchers; that this is not the case seems odd). Local
IRBs might elect, for example, to convene an executive session to consider
individual proposals, in the interest of protecting the privacy of
individual researchers or participants. But when considering administrative
issues, such as rule changes and appointments to the IRB, it should be
apparent to the local IRB and other regulators that the more input from the
participants in the system, the better the system will be, or at least the
better the level of acceptance will be.
Regulations are highly subject to change and are not subject to much
scrutiny about the purpose or justification for them, particularly on the
part of those trying to follow them. Scrutiny and talk about purpose might,
of course, lead to de-regulation, or de-escalation in regulation. The
purpose of ideology critique is to reclaim organizations as
“social-historical constructions,” investigating the formation of the
organizations and ways they are sustained and transformed “by processes
both internal and external to them” (Alvesson & Deetz, 1996, p. 200;
see also Giddens, 1979; Deetz, 1985). Institutions comply with the federal
system voluntarily, except in the relatively limited number of cases
involving federal funding.{204}*
Clinical researchers were the first to be regulated (and now many times
fall under the purview of the FDA){205}*.
But when social scientists were (or allowed themselves to be) co-opted by
the system, the regulatory process, in many ways, became absurd.
The IRB system has been allowed to go too far, with respect to minimal
risk{206}* studies in particular (Flyvbjerg, 2001). This is
evidence of the second theme presented by Alvesson and Deetz (1996),
universalization of managerial interest. This case involves the missions
and cultural reproductions of operators within government bureaus,
departments, agencies, legislative bodies, and the professional press, with
their near myopic focus on clinical trials{207}*
(see Figure 1,
p. 50). This exclusion of the needs and interests of social science
researchers (because of the seemingly singular focus on clinical trials, or
the unfortunately prevalent idea that somehow one set of rules can work for
both clinical and social research) creates many problems. Regulators, if
they have conducted research and are typical researchers, have mostly been
involved in clinical research (this is also perhaps a bias/blindness on the
part of those hiring/appointing regulators; see also NIH, 2001, Jun 26, p.
1-6; and NBAC, 2001, re: percentage of clinical studies reviewed by IRBs).
The rules are made with clinical trials in mind.{208}*
Alvesson and Deetz (1996) suggest “management is ascribed the superior
position in terms of defining the interests and interest realizations of
the corporation and thereby of wide segments of the population” (p. 200).
Guided, at least loosely, by the general federal rules, local institutional
regulators through their interpretations, tell researchers in highly
specific ways what is important in the conducting of research, i.e.,
when they are in compliance with the details of IRB interpretations. The
local IRB, via the discursive formation that constitutes the system
(Foucault, 1972) tells researchers what constitutes “appropriate” human
subjects research, and when they are doing it. (SINS) STRUCTURES,
INSTITUTIONALIZATIONS, NATURALIZATIONS, SIMULATIONS contribute to the
attitude that IRB processes are necessary, self-evident, inevitable or
unchangeable; the idea that local IRBs make rules consistent with federal
rules; the belief that conflict, dissention, and anarchy are negative
(often inherently evil) forces; and the notion that in spite of what seem
to be highly unreasonable and nonsensical processes within the system, that
somehow the overall system is reasonable and sensible. We tell ourselves
that there must be something we don’t know, or aren’t able to understand (i.e.,
we rely on mysticism). This is another byproduct of a simulated existence, i.e.,
we confront what is clearly, locally “really” stupid, and suggest it is
stupid because of our own ignorance. That is likely true only some of the
time.
The emergence, dominance, and persistence of instrumental reasoning,
specifically technical rationality, “governed by the theoretical and
hypothetical and focusing on control” (Alvesson & Deetz, 1996, p. 200)
an “unswerving application of [science] methods without reflecting on
knowledge-constitutive interest”, i.e., “[taking] the historical
traces of suppressed dialogue and reconstruct[ing] what has been
suppressed” (Habermas, 1971, p. 315),{209}*
is readily apparent at the institutional level in the IRB system. This
instrumental reasoning is apparent in the way the rules are written, by and
for whom. It is also apparent that the interests of the participants are at
least occasionally supplanted by alternative purposes, i.e.,
protection of funding sources, avoiding litigation, political posturing,
etc. As mentioned, most of the activity is paperwork that clearly dominates
the system and the people (researchers and regulators) participating in it.
These participants go along, most generally without much resistance, as if
this paperwork, original or revised, produces the desired (and/or an
undesired, and/or any or no) result. Adopting and following this process in
great detail “protects” us from actual engagement in issues surrounding the
purpose of what is being done. Treating the procedures as
handed-down, customary, “self-evident” routines “unburdens” researchers
from the necessity of directly examining issues in the use of human
subjects and issues in steps to protect them. (For customs as unburdening,
see Schutz & Luckmann (1973, p. 298).
Historically, much more two-way communication is generated in the local
IRB/individual researcher interface than in either the federal
regulator/local IRB interface or the federal regulator/individual
researcher level. The two latter interfaces are almost exclusively one-way,
and the large majority of interfaces involve mostly impersonal
communication, unless some anomaly/tragedy occurs (a death and/or
sanctions, as examples). Researchers don’t interact and correspond with
federal regulators until/unless a “problem” arises. There is no natural
interface in the system design.
Testimony given during the U.S. House (1998, Jun 11) hearing revealed
the OPRR didn’t know the exact number (could not offer a guess to within
several hundred) of IRBs operating at that time, maintained no central
registry, nor a clearinghouse of complaints or any other paperwork
institutions are required to file with the federal government. Only the
assurance documents (and then only if federal funding is involved) were
maintained by the OPRR. Ellis indicated, “We do currently have a list of
the name and address of every IRB that’s under our authority” (U.S. House,
1998, Jun 11, p. 65). The OPRR had no legal reason to keep such a list;
they have no jurisdiction unless federal funds are involved. It may have
been the posture of the OPRR that (given the overwhelming majority of
institutions follow the rules voluntarily) there was no organizational
compulsion to insist upon a registration process for IRBs. However, this is
one of the first things that the new incarnation of regulators, i.e.,
the OHRP, changed after the (somewhat superficial) structural changes to
the system in June 2000 (see IRB Registration and Federalwide Assurance
[FWA] Questions and Answers at http://ohrp.osophs.dhhs.gov/humansubjects/finreltn/faq.htm,
accessed May 25, 2002). The OHRP now has a highly articulated, complex, and
mandatory registration procedure for IRBs and discussion of certification
or accreditation activities are gaining prominence in the discourse (see
Southwick, 2002, May 3; Human Research Subject Protections Act of 2002; and
Southwick & Monastersky, 2001, Jun 1).
Reliance on instrumental reasoning, aiding dominant groups’ ability to
invisibly accomplish their ends, is described by Habermas (1971) as
instrumental technical reasoning. “The understanding of meaning is directed
in its very structure toward the attainment of possible consensus among
actors in the framework of a self-understanding derived from tradition” (p.
310). Instrumental technical reasoning is apparent in the way regulators
(appear to view) themselves and their power, responsibilities, authority,
etc. For example, by offering training, as suggested by Ellis when he was
head of OPRR (and now Greg Koski, director of the new OHRP) these
regulators create and maintain the “proper” view that is to be held
(somewhat like textbooks, stylebooks, handbooks, guidelines, curricula,
summaries, reviews, etc.). Koski, as quoted in Brainard (2000, Jul 21, p.
A21), says, “Most clinical researchers lack formal training about how to
conduct clinical research in an ethically appropriate manner, and should be
provided with such training.” Koski’s comment demonstrates this creation
and maintenance of the “proper view,” i.e., through his use of
“formal training” in order to act in an “ethically appropriate
manner.” He did not say that clinical researchers don’t know how
to act ethically, or that they are not acting ethically, but that
they lack formal training about how to act ethically, and that the OHRP can
provide such “knowledge” (begging at least two questions: whether
researchers know or not, and whether the OHRP knows or not). The federal
views become institutionalized in this way (Koski’s “right” to define what
the problem is and who has it, and the solution and who has that) and his
(organizational) right to establish and maintain the (only) “proper” views
and behaviors (demonstrating the managerial/regulator imposition set forth
in Alvesson and Deetz, 1996).
Based on their interpretations of the federal provisions (not only each
individual provision, but also the view of the role of the federal
government/regulation overall), institutional regulators create lists of
approved standardized forms, templates for forms, and provide even more
highly specific (and more-rigorous-as-virtuous) details about what must be
stated on the forms.{210}*
In other behavior at least in part designed to standardize, i.e.,
define and set and police standards in this case, the National Association
of IRB Managers has offered the IRB Management Certification Examination
since 1995, indicating the development of tools to standardize the
thinking, and subsequent certification of IRB members themselves (see
National Association of IRB Managers [NAIM, 2001], or http://www.naim.org/cert.htm,
accessed May 25, 2002; see also Southwick & Monastersky, 2001, Jun 1;
and Southwick, 2002, May 3). These are highly presumptuous attempts to
control the process. Hegemony is a process of infiltrating common sense
(after having devalued it in the hegemonic discourse of science, as
mentioned) and becoming part of the way we construct reality, the way we
view the world, understand ourselves, experience needs, or get certified
(Alvesson & Deetz, 1996; Gramsci, 1971). We now “need” (and actively
seek) IRB approval as researchers (and regulators, based on the previous
paragraph, may be actively seeking approval too, i.e.,
certification), however, the “real” the need is for people to be treated
ethically (i.e., fairly, safely honestly, etc.). This creates a
discrepancy between what is and what could have been or what could
be, regardless of how wide or narrow the gap between (the size of which
can’t be known). As postmodernists might argue, if we leave the future to
the future, it might be better than anything we can engineer, or delude
ourselves into believing we have designed, strategically planned, or even
affected. The need for IRB approval is not inevitable; death is inevitable;
filling out forms is minutiae, and not inevitable.
There is usually a difference between the interests of the
dominators and the dominated, even if the dominated don’t feel dominated
(this speaks to the depth of the SINS) or are not able (or willing) to
admit/articulate the problem. It is unreasonable to presume to know
what the dominated want or need, or even who they are, but to argue a
discrepancy can be seen, as in this case, is reasonable. Regulators, in
order to complete their job mission, attempt to standardize the process of
gaining approval for human subjects research. Researchers, though not continually
re-inventing wheels, are inherently looking for something that hasn’t been
done before (perhaps describable as “neo-standard”). In fact, the size of
the discrepancy, (an unknown value), might be viewed as an indicator of the
(necessarily estimated) depth of oppression (at least in part attributable
to SINSful acceptance). Lack of visibility, as Foucault (1980) would
suggest, enhances the effectiveness of power (see also O’Connor, 1979,
especially p. 226).
When visible, motives, reasons, damages, and other objects can be
described and critiqued, and debated and changed much more readily. Open
conflict about the differences in those motives, reasons, and points of
view are more likely to occur; new decisions can be made (as well as new ways
of going about making them), old ones abandoned, usually with plenty of
creativity, blame, satisfaction and other by-products of the activity
readily apparent.
A sterile world of sameness is, of course, the great risk in hegemonic
success. “Cultural diversity is dissolved in the acid bath of the core
corporate values” (Willmott, 1993, p. 534), but “management control
strategies are seldom fully successful … Resistance and some level of
cultural diversity normally prevail” (Alvesson & Deetz, 1996, p. 201).
Telecommuters and others who take unorthodox approaches to work and
organizing{211}* have survived, for example; some have even become
more prevalent over time. According to In-Stat (2001, Feb 24),
roughly 24% of the U.S. workforce is estimated to telecommute some time
during the week in 2001. This works out to be more than 30 million at-home
workers. In-Stat expects this percentage to increase to 28% in 2004,
growing to nearly 40 million workers. Research diversity, including the
ways of approaching the work of research, is important (in my own hegemonic
discourse). The system we have installed, however, works against diversity
(as many similar systems do). Trends toward homogeneous processes are
particularly important to avoid in educational pursuits where developing
“new” ways of going about or looking at things is often a major purpose of
the activity.
At the workplace, people now “find” their physical (company softball
and/or training facility), social (company parties, support groups),
political (union dues funneled to PACs, the boss’ bumper sticker, and more
overt means),{212}*
financial (paychecks, insurance, retirement plans, and assorted carrots),
and even spiritual lives (company mentors, motivators, inspirational
corporate counseling). This creates a “total” environment, much like
cultists of other sorts, that serves to strengthen the bonds of control
within the system, and weaken “outside” links with family, community, etc.
In this way, management control is not limited to our professional
lives. Deetz (1992) suggests “the decline of vision, hope, and community in
politics has paved the way for management ideologies and practices” to fill
what he calls “the vacuum” (as discussed p. 209 of Alvesson & Deetz,
1996). If a person’s life is/contains a vacuum, I suspect these forms of
control work even better than if a person has a strong identity in place.
This is relevant to the student researchers, as mentioned previously (see
p. 202 and p. 247-248, herein). I don’t mean to suggest students’
lives are any more or less vacuous than other lives, however, students are
not as likely to have a strong identity as researchers the first few
times they make an application to an IRB. This “control” idea also bears
strong resemblance to Theory X (McGregor, 1960) thinking, implying
challenges to the integrity, recall, and humanity of all researchers,
students and others. Student researchers, therefore, while being exempted
in some disturbing ways, are, perhaps, more abused by (an unclear,
contradictory, unreasonably convoluted) the system than application savvy,
politically astute researchers, ones with IRB experience, and who have
perhaps more confidence, ability, and willingness to circumvent what they
have concluded are tedious and pointless, required but unnecessary
processes (see Manning, 1978; see also Overland, 2002, Feb 19, re:
corporate circumvention of IRB processes via conducting trials in other
countries).
Corporate cultures and government/institutional structures, when viewed
as text, suggest culture members become “readers” who bring their
experiences and awareness of other texts and other cultural forms, their
histories and their own moods and personalities to their reading of texts.
They “enter into the text, changing its nature and reproducing it as they
consume it” (Linstead & Grafton-Small, 1992, p. 344). Reality and truth
are liquid rather than solid, it appears, logically alarming some
(positivists, propagandists, and itinerarians, as examples) who believe it
is possible, necessary, and/or desirable to control the routines of others.
Whether regulators label a situation as “problematic” or an
“infraction,” or describe a situation as “systemic” or “local,” whether
they impose sanctions or conduct investigations, whether the U.S. Congress
passes a bill to supplement current laws with the force of more law or not,
and whether a researcher considers a study doable or not in the face
of IRB rules are examples of the outcomes of “text reading.” Variance in
readings reveals underlying values in conflict (notions about the rights of
individuals versus the common good, for example), the range of
interpretations brought to any given text, and even which texts one will
“read” at all. And, in matters of academic free inquiry, federal regulators
appear to adopt a “least restrictive environment” position. But this policy
does not hold at the institutional level of interpretation (see AAUP,
2001). It appears odd that institutions do not embrace this basic tenet of
administrative law (i.e., “least restrictive” as preferred over
“more rigorous”) as a win-win possibility among IRB members, researchers,
and participants in research.
Dialectical tensions in the system. Work in hermeneutics (see
Alvesson & Sköldberg, 2000, particularly p. 52-109) is useful in
considering the IRB system in its very complex social context, particularly
in helping to formulate questions for the purpose of continued and enhanced
whying. For example, framing the tensions between institutional compliance
with regulation (the event) and competition with other research
institutions for funding (the context), and attempts of institutions and
researchers to negotiate the political environment, avoid litigation and
sanctions (or explain either or both and to balance all these things and
more) lead to many important questions about the IRB system of protections,
focusing on those questions about actual protection of human subjects,
such as conflicts of interest, personal and institutional funding pressure
and its effects on researchers’ strategies, considerations about the
qualifications (and/or accreditation) of IRBs and/or the members of them,
etc. (See Andrews, 2000, Mar 10; AAUP, 2001; ACHRE, 1995; NBAC, 2001;
Blumenstyk, 2000, Nov 6; 2001, Apr 26; Blumenstyk & Wheeler, 1998, Mar
20, Southwick & Monastersky, 2001, Jun 1; Southwick, 2002, May 3; and
Human Research Subject Protections Act of 2002.)
Sanctions against the University of Oklahoma-Tulsa Campus issued in June
2000 were related to concerns about patient safety, and brought about as a
result of an outside consulting firm’s report and a whistleblower’s
efforts. This situation involved a clinical study in which the principle
researcher was also the inventor (and the manufacturer, see below) of the
vaccine used in the study, as well as being a doctor to the patients
participating in the study. (None of the patients referred to this
individual as a researcher, as anything other than “doctor,” which is
indicative that the consent was “really” as confused as it was informed.)
In addition to this conflict of interest, the situation on that campus was
incestuous in that the doctor/researcher in charge of the study, Michael
McGee, had also, and without FDA or any other authorization, manufactured
(without the required outside monitoring) the vaccine that he then
administered to patients. The events were conspiratorial in that McGee and
IRB chairman Daniel Plunket misrepresented to patients and members of the
IRB the reasons for an earlier suspension of the study, i.e.,
Plunket, who later retired rather than resign or be fired, didn’t inform
the rest of the IRB or the president of the Tulsa campus, the vice
president for research, the federal government, or anyone else about the earlier
suspension of the trial. Plunket and McGee kept the issue between
themselves.{213}* It took one (rather courageous and persistent)
whistle-blowing nurse to take the information about the mis-management of
this clinical trial to the federal regulators, who issued sanctions in June
2000. In all, four heads rolled including those of the principal researcher
in charge of the trial, Michael McGee, who was also vice chairman of the
Department of Surgery; Harold Brooks, dean of the college of medicine at
the University of Oklahoma-Tulsa; Edward Wortham Jr., the director of the
Office of Research at the OU Health Sciences Center; and Daniel Plunket,
the chairman of the college of Medicine’s IRB and senior associate dean for
clinical affairs (see Winslow, 2000, Jul 22). Winslow (2000, Aug 1)
reported about the aftermath of the sanctions and firing, specifically the
level of confidence patients had in the vaccine and the doctor/researcher.
There were also letters to the [Tulsa World] editor from patients
during this time, and Time (Lemonick & Goldstein, 2002, April
22) featured the events in a cover story). None of those rules worked in
this case. In spite of this rather massive rule failure,{214}*
OU officials then created new rules and processes such as new layers
of checks, including unannounced spot checks; the appointment of a new
central research compliance office; a 24-hour hotline allowing anonymous
callers to report violations; mandatory employee training about compliance;
and, similar to the institution-to-federal government assurance document,
an oath, processed from individual to institution. And, officials stated
the obligatory cliché, i.e., the new rules are designed to,
according to the Tulsa World (Kerr, 2000, Jul 21) “ensure an
incident like this wouldn’t happen again,” as was heard in similar form
from officials in response to Nuremberg, Tuskegee, University of
Pennsylvania, Duke, Johns Hopkins, and so on and on. OU President David
Boren said it was “good” that the incident happened when it did (before
more than one percent of OU’s research was conducted in Tulsa), and talked
about becoming, not (necessarily) the model of human protection, but
the model of compliance, “I do think we will see the day when these
procedures will be used by other universities as a handbook” (Boren, quoted
in Gillham, 2000, Jul 23; for details of other cases, see Andrews, 2000,
Mar 10 regarding the Gelsinger case at the University of Pennsylvania; Cho,
1997, Aug 1, a discussion about the increasingly dubious ties between money
and academics/clinical trials and the effects on rules; Brainard, 1999, Oct
1, about sanctions at the University of Chicago and the resignation of a chancellor
there). Rather than a handbook, the events at OU are being used as a case
study in how not to conduct studies (see Lemonick & Goldstein,
2002, Apr 22; and Southwick, 2002, May 3, re: testimony about the case in
Senate committee hearings).
Tensions about priorities. IRBs focus on minutiae,{215}*
even sub-minutiae and concerns well outside their purview (Brainard, 2000,
Mar 17). Beyond those used by Brainard, examples of minutiae, myopia, and
expansionism I have learned about directly (i.e., in “real” life)
include board member “suggestions” about the title of studies, whether or
not the IRB has “jurisdiction,” and whether or not a study is “valid,” (a
study which may employ methods with which the IRB member has little or no
familiarity, but who may assume s/he does, and in a manner of confusing
authority with superior knowledge, apply “clinical trial” standards. And it
appears IRBs focus too intently on (representations of) studies (i.e.,
protocols from researchers) requesting permission to conduct studies
involving no treatment{216}*
(see Brainard’s study, “An inside look at how a university tries to protect
human subjects,” 2000, Mar 17, in which he observed and interviewed Duke
officials after the sanctions there and came to conclusions similar to
those mentioned here). Many university scholars and government officials at
an August 2001 conference sponsored by the DHHS expressed concern that
scrutiny of researcher’s financial interests would fall to IRBs. “These
panels, which are required by federal regulations to monitor the risks of
proposed research involving human subjects, are widely viewed as overworked
and understaffed. Only about a fourth of the IRBs surveyed by the inspector
general at [DHHS] routinely review conflict-of-interest issues” (Brainard,
2001, Feb 16, p. A33; see also DHHS OIG 1998b, 2000b; Mangan, 2000, May 19,
and 2000, Oct 30; Blumenstyk, 2000, Nov 6; Okie, 2001, Aug 6). I would add
that the local IRB members would be under undue pressure, i.e., they
would be under pressure not to stand in the way of funds, i.e.,
disapprove certain studies. This is a situation in which “local” may be too
close to home. It is not the best system that could be developed if the
goal is human subjects protection. But, as mentioned, that goal is mostly
superficial in the way processes are developed.
Conflicts of interest tensions. Large financial rewards
have become more and more a part of the research environment. Conflict of
interest policies have evolved, including disclosure requirements, at
research facilities in universities and in professional journals. In May
2000, Harvard University drew fire for “considering whether to ease some
financial restrictions in its conflict-of-interest policies” (Mangan, 2000,
May 19, p. A47; see also Mangan, 2000, May 26). According to Mangan, the
new guidelines were perhaps necessary to retain researchers. Harvard’s
policy, according to many of those quoted in Mangan’s articles, is the
strictest policy in the country. Sheldon Krimsky, a professor of urban and
environmental policy at Tufts University, said “Harvard seems to be moving
in a direction that’s going to lead to more conflicts of interest and
greater secrecy in research” (as quoted in Mangan, 2000, May 19, p. A47).
Krimsky continues, saying if Harvard relaxes its policy, it would be
“lowering the bar at a time when we’re getting more information that
indicates that close ties between universities and industry have adverse
consequences to the health of science and to the public interest” (as
quoted in Mangan, 2000, May 19, p. A47).{217}*
Mangan also states that Harvard’s consideration to ease its policy is
“responding to competitive pressure” and surmised that the death of gene
therapy patient Gelsinger in 1999 “may cause some schools to tighten their
controls over sponsored research” (2000, May 19, p. A47).{218}*
By October, in another article by Mangan (2000, Oct 30), the debate
about conflict of interest had escalated, leading to the formation of a
panel (the effort led by the president of the Association of American
Medical Colleges) to “help forge a consensus on how schools should deal
with the problem” (Daily News section). The president of the
Association of American Medical Colleges, Jordan Cohen, said, medical
schools need to agree on a consistent strategy for confronting financial
conflicts of interest that are steadily eroding public confidence in
clinical research (and taking social scientists with them, incidentally and
unfortunately). “We risk great peril if we fail to respond to the growing
perception that financial conflicts of interest have gotten out of control”
(Daily News section). (Notice the use of the term “growing
perception” rather than an acknowledgement that financial conflict of
interest is a “real” problem.){219}*
Cohen states “A conflict of interest is like potential energy: it has the
capacity to cause something to happen, but until unleashed, [it] is simply
a lurking presence” (as quoted in Mangan, 2000, Oct 30, Daily News
section).{220}* The debate has further escalated to the level of the
OHRP proposing rules for “institutions, clinical investigators, and IRBs to
consider when dealing with issues of financial interest and human subject
protection” (see OHRP, 2001, Jan 10; available: http://ohrp.osophs.dhhs.gov/nhrpac/mtg12-00/finguid.htm,
accessed May 25, 2002; also see Brainard, 2001, Feb 16.) However, more than
half the states are considering or have already passed laws designed to
eliminate barriers to collaboration between universities and private
companies, with “conflict-of-interest fears [taking] a back seat to
economic development” (Schmidt, 2002, Mar 29, p. A26).{221}*
Many of these states made sizable expenditures on research universities,
tightening up the relationship between “learning” and “learning to profit.”
Ricoeur (1971) states “In the same way that a text is detached from its
author, an action is detached from its agent and develops consequences of
its own. This autonomization of human action constitutes the social
dimension of action” (p. 541) not only because organizations are produced
by numerous actors whose individual roles are not distinguishable, but also
because “our deeds escape us and have effects which we did not intend” (p.
541). This includes what we write. These effects are apparent in the IRB
system, as in Cohen’s comments above, i.e., conflict of interest
issues are a “lurking presence” constituted by “growing perception[s].”
These ideas resonate with those from the NIH: “Why are there concerns?
Concerns are raised when financial considerations may compromise or have
the appearance of compromising an investigator’s professional judgment
and independence in the design, conduct, or publication of research” (NIH,
2000, Jun 5, Notice OD-00-040, emphasis added). These concerns don’t
mention human subjects specifically, and point out the “appearance of
compromising” as (equally) important.{222}*
The power of atrocities. At the institutional level,
changes and interpretations are rarely driven, at least directly, by
atrocities or sanctions. Both, thankfully, are rare; unless the system
changes dramatically, most institutions will never experience them
directly. However, many institutions are affected when atrocities or
sanctions occur anywhere (see comments after Duke sanctions in Brainard,
2000, Mar 17, especially Duke associate professor of medicine and philosophy
and bioethicist Jeremy Sugarman, who is quoted, “I got e-mail every day
[after the suspensions were announced] from people at other universities
who said, ‘There go we, but for the grace of God’”). Self-regulation,
normalizing cultural creations, conceding, and behavior too similar to
mooing follow atrocities and sanctions, with institutional regulators (with
managerial superiority) often being the first over the cliff, followed very
shortly by the compliant non-whying herd of researchers.{223}*
In the absence of (almost exclusively medical) atrocities, tragedies, or
sanctions, local institutional regulatory drive must (necessarily) come
from local experiences. Changes in personnel of an IRB itself or
higher-level managers/administrators may create change. New people have new
(if not necessarily good ideas, in the view of some; ideas are perceived as
bad if they don’t support the status quo, of course), and incumbents
possess a generalized anxiety toward change and difference, a “natural”
desire to maintain their way of life, to avoid change and the unknown (a
form of psychological inertia perhaps) regardless of the organization
examined. This contributes to the static nature of IRB systems (and plenty
of other bureaucracies). This is not to say that resistance to changing
rules means there is no activity in the IRB system, rather that the
activity leads (nearly) nowhere (see next section). As Brainard
(2000, Mar 17) suggested, it is easier, and safer, to focus on wordsmithing
consent forms than to attempt to answer larger—and
self-reflective—philosophical questions about what they as regulators are
doing overall, or what they are entitled to do in any given instance. As
Adorno (1989a) writes, “the completely reified and mediated is a sanctuary
from immediacy and life” (p. 130; see also Schutz, 1973, p. 298, and
footnote # 209, p. 195 herein for a relevant statement from Habermas).
SINS-ful sameness. It might be expected that the IRB system
would produce high levels of standardized, often strategized, formulaic
discourse that doesn’t change in meaningful ways the “real” world of human
protection. As shown in various examples herein (including Glenn’s
discourse, both in the writing of the bill and the act of defending the
virtues/establishing the need for it, and, further, in the result of the
efforts—the bill never got out of committee, i.e., any result
related to the purpose of regulations was nothing much, though other
benefits may be said to have been gained. The discourse may have enhanced
Glenn’s political capital, or reduced litigation potential for
institutions, including many government agencies conducting research. But
to a much lesser extent were protection provisions affected.{224}*
(SINS) STRUCTURES, INSTITUTIONALIZATIONS, NATURALIZATIONS,
SIMULATIONS (such things as the institutionalization of the application
process/forms, the naturalizations associated with “the way we do things”
at the IRB, etc.) operating in the local regulatory system have worked to
create sameness.{225}*
This standardization may be beneficial to the IRBs by alleviating, if not
eliminating, the need to fully engage in protocol reviews because
eventually (maybe even rapidly) the proposals become intensely formulaic as
researchers adopt the IRB rules, specifically, conforming to the
application process (a comfortable “default” as described by Habermas,
Schutz, and Adorno and mentioned on the preceding page). Conformity, by
definition, breeds sameness. I would suggest it is possible, even likely,
that applicants whose proposals conform, i.e., reflect the formula,
are more often and more quickly approved, and those applicants whose proposals
do not conform are subjected to more scrutiny and more often rejected than
“standard” applications (this is the case at the OUIRB, though if my own
somewhat unorthodox applications had been approved it would have been
possible for me to provide more precise observations). This is, it might be
argued, a form of “pre-approval” bias, albeit with no guarantees. Often
IRBs, in the revise-and-resubmit routine, “guide” the researcher to
conformity and cookie-cutter sameness, if the researcher doesn’t find it on
her/his own{226}* (see Brainard, 2000, Mar 17 and White, et al., 1995,
for discussion of typical reasons for IRB rejection; see footnote # 233, p.
226). The process, particularly of the researcher’s application, becomes an
IRB review of the structure of the proposal rather than a review of the
treatments described in the proposal.
Other effects. Other local effects include the following as
potentialities of the system, given the federally mandated structure and a
university based IRB’s expectable membership (at least historically):
activities related to the avoidance of lawsuits; previous experience with
the applicant, topic, and/or methods; and the IRB member’s own view of
power. A generally absent yet potentially important “local effect” is
researcher insistence on a better system; resistance of a simultaneously
intrusive and impotent one. Demanding a voice in the process (i.e.,
legitimation rather than marginalization) and/or participating in passive
rebellion rather than in an inappropriate system are rather obvious choices
not often selected (perhaps in a socio-professional version of survival of
the fittest, i.e., to win, one must play the game; dissenters, by
definition almost, don’t “fit in” as they are playing a different game, too
often whying about rules, making trouble, etc.){227}*.
Adorno (1989b) says it this way:
If we were looking for an
ideological justification of a situation in which men are little better
than cogs to their own machines, we might claim without much exaggeration
that present-day human beings serve as such an ideology in their own
existence, for they seek of their own free will to perpetuate what is
obviously a perversion of real life. So we come full circle. Men must act
in order to change the present petrified conditions of existence, but the
latter have left their mark so deeply on people, have deprived them of so
much of their life and individuation, that they scarcely seem capable of
the spontaneity necessary to do so. (p. 275)
Power sources. Legitimate power, it is argued in the
definition provided by French and Raven (1959), is exercised only to the
extent that control is acceptable to the controlled. This idea is not inconsistent
with Foucault (1980), Giroux (1988), Hall (1986, 1989), and Weber (1947,
pp. 324-341), who was an original source on this sort of definition – he
called it “legitimate authority” rather than “power.” These notions about
power are particularly relevant in the IRB system, as this system is
heavily reliant, on both the federal and institutional levels, upon
voluntary compliance{228}*,
and most research is unsupervised (see “lack of direct oversight” concerns
in DHHS OIG 1998b, 1998d; GAO, 2001; NBAC 2001, and others). This is
another example of the way power operates as a relationship (Foucault,
1980), i.e., power must be given before power may be possessed.
Researchers have given power to the system. And, if one is to follow the
liquid and local line of argument, specific researchers give specific
power(s) to specific regulators at specific times. They don’t have to. They
do. They may (falsely) believe they have no choice. But, “really,”
they, we, do.
Impacts of Various Dogma on Current System
Scientific management theory. Scientific and classical
management theories as mentioned previously, comprise some of the earliest
attempts to understand the complexity of organizations. The works of
Taylor, Fayol, and Weber are examples. Taylor (1919/1947) first set forth
his views of scientific management as early as 1911. Fayol and Weber
published work about the same time, although not available in English until
the 1940s; Taylor’s work wasn’t compiled and published until 1947. The IRB
system was being formed about the same time as these works were becoming
available (the AEC was formed in 1947). These theories share a common
tenet, namely that organizational efficiency is determined by the efficient
design of work and the organizational structure, not inconsistent with
Lyotard’s (1984) performativity ideas. In looking at the historical
developments, these classical theories may be useful in gaining better
understanding of the social context that existed at the time of IRB
development, and the (SINS) STRUCTURES, INSTITUTIONALIZATIONS,
NATURALIZATIONS, SIMULATIONS we enjoy/endure today.
Taylor’s (1919/1947) scientific management system suggests, among other
things, that there is “one best way” to perform any job (part of the root
ball of the (SINS) STRUCTURES, INSTITUTIONALIZATIONS, NATURALIZATIONS,
SIMULATIONS of standardization), and that employees should be selected in a
measured way (based on test performance or longevity, for example). In the
IRB situation, not only federal rules about how many people and what
categories must be maintained on the board, but local traditions may exist,
such as having all the colleges represented on the board, or other “local”
criteria that may become viewed as the “one best way” of “doing regulation”
(see Sacks, 1970).
Another element of the scientific management system is that labor should
be divided so that managers plan and workers follow those plans, i.e.,
an authoritarian hierarchical system. IRB members operate, as do many regulators,
with don’t-ask-why, do-it-because-I-said-so, I-didn’t-make-the-rules,
it’s-the-way-we-do-it-around-here attitudes. Rules may be interpreted in
sometimes vastly differing ways and applied in still other ways, suggesting
the rules themselves are liquid, but the pattern of the process is
more solid. For example, one application may take much longer to consider
than a similar one for no apparent reason, (knowledge gained from numerous
personal conversations during the past few years with fellow graduate
students and professors) yet the process of making the two
applications is virtually identical. And, sometimes even similar content
may be in present in two or more applications treated differently by an
IRB, i.e., depending upon a particular reviewer’s view on a
particular day, decisions may be inconsistent over (rather short periods
of) time. These inconsistencies may also be due, in part, to the attitudes
of both the regulators and researchers that (at least occasionally) they
are doing something that seems “nonsensical” or contrived (a
we-have-to-do-this-even-though-it-seems-silly feeling even regulators must
get from time-to-time). It is difficult to be “logical” or under a blanket
of nonsensical circumstances/circumventions (creating/contributing to circumvisions
perhaps?).
Fayol’s (1916/1949) general management theory outlines 14 fundamental
principles. Those relevant to this study are his notions of authority and
responsibility, discipline, order, and perhaps most relevant, the
subordination of individual interests and the development of esprit de
corps (or more contemporary monikers: “production teams,” or “corporate
cults”). Faculty members encourage students to write applications with the
goal of gaining approval—that is the focus of discussion in methods
classes, both qualitative and quantitative. The discussions are not focused
on the “real” purpose but on the process and strategies for
enduring/accomplishing it.{229}*
This is unfortunate for everyone concerned, with the possible exception of
the IRB members themselves. They enjoy standardized applications and even
(an insidiously) comfortable standardization of studies,
contributing to job conditions that are simpler (and safer, politically and
legally.) “Standardization” may be a preference of many people (not just
regulators, but faculty, too, for example). But I would suggest this is an
example of hegemonic thinking. Many of these same people might also
acknowledge, if they “really” thought about it, that almost always
standardization of a (nonsensical) process is not a superior option to
exemption from that process. In qualitative methods classes, for example,
it would be preferable (and consistent with the purpose of IRB
regulation) for many of the IRB proposals required of students (and
subsequently evaluated by faculty) to be eliminated. Exemption is far
preferable to standardization, I argue, for some Very Big reasons (avoiding
intellectual suffocation not the least among them) and some smaller reasons
(attempts to standardize the non-standard world can be a Very Big pain in
the ass, though dismantling a standardizing structure is an Even Bigger
One). Of course, thinking un-hegemonically is difficult, troubling, risky,
and exhausting.{230}*
Process “templates” are a manifestation of managers’ (regulators’ in
this case) co-opting of volunteer “subordinates” (researchers,
particularly) to take part in the system, i.e., to participate in
the (SINS) STRUCTURES, INSTITUTIONALIZATIONS, NATURALIZATIONS, SIMULATIONS
of the regulatory (and as a gift, dominant) “scientifically managed”
cultural reproduction. Illustrating this is the behavior of the many
graduate students and faculty members who circulate advice about what sorts
of studies should be avoided (or adopted) if reasonably quick IRB approval
is to happen. Their advice is based on their own individual histories with
the IRB, what they feel they understand of the IRB policies, and rumor,
too.
One of the (SINS) STRUCTURES, INSTITUTIONALIZATIONS, NATURALIZATIONS,
SIMULATIONS in the IRB system is the belief that a “best for all” condition
exists, á la Taylor, and Plato. (I would suggest that the idea of a
“common good” as possible is required to drive a system toward
standardization, i.e., “template fever.”) Even regulators themselves
indicate that if “template fever” runs too high, the template may add
symptoms (demonstrating, incidentally, that regulators can admit that even
templates have points of diminishing returns) rather than operating as a
cure for a sick system. (In this “fever” scenario, it would follow that the
system could occasionally become delirious). The OHRP, in its letter to
Johns Hopkins University (OHRP, 2001, Jul 19) outlining the reasons for
sanctions in the wake of the death of a healthy volunteer in an asthma
trial, stated, “OHRP is concerned that the boilerplate informed consent
document is difficult to understand and contains information that may be
irrelevant for certain research projects” (p. 8) yet on the previous page,
language used by the OHRP seems to contradict, or at least point out the
difficult dialectical tensions between providing enough but not too much
information on the consent form.{231}*
In particular, OHRP (2001, Jul 19) suggests that the IRBs “encourage
investigators to limit the length of informed consent documents, and as a
result, important information is being excluded;” “OHRP again strongly
recommends that the informed consent document boilerplate used by the IRBs
and checklist be modified to include the additional elements at 45 CFR
46.116(b);” and “OHRP is concerned that the informed consent documents
approved by the IRBs often appeared to include complex language that would
not be understandable to all subjects” (p. 7). A similar situation, with
respect to limiting the length of informed consent forms to something
substantially shorter than the instructions provided, exists at the
University of Oklahoma, as mentioned (see University of Oklahoma, IRB
application form, http://research.ou.edu/Forms/index.htm, accessed May
25, 2002.). These statements do little to clarify what Johns Hopkins should
do, but they speak volumes about the difficulties in delivering informed
consent in the “real world”{232}*
and the inconsistency in the simulation, born from dialectical tensions in
the “real” world. In the OHRP letter to Johns Hopkins (OHRP, 2001, Jul 19),
regulators indicate that the use of a boilerplate and complexity of the
form may have contributed to the problems related to the death of a
volunteer, that the informed consent form “appeared to include complex
language that would not be understandable to all subjects.”
Not only are the subjects of research perhaps unable to understand the
complex language of consent forms, similarly, researchers may not, as
mentioned, understand the complex language used in or the structure of the
process. For example, the instructions OU provides to explain the informed
consent document and how to successfully negotiate it are substantially
longer than the “recommended length” of the informed consent document itself.
While it is understandable that this would occasionally be the case in the
world, it is particularly troublesome here. Given that many of the
instructions include lengthy passages about actual verbiage that must be
included on the informed consent document (an example of wordsmithing, as
has been mentioned previously, see footnote # 18, p. 16, p. 212, and p.
221), it appears the “instructions” are impossible (or nearly so) to
follow, i.e., to include all the required information produces a
document longer than the recommended length, found in the same set of
instructions.
Finally, Weber (1947), in reporting his observations of bureaucratic
functioning, suggests the “ideal” organization (i.e., a “real”
bureaucracy) in the eyes of practiced bureaucrats has six basic features: a
clear hierarchical system of authority; a division of labor according to
specialization; a complete system of rules regarding the rights,
responsibilities, and duties of personnel; exhaustive procedures for work
performance; impersonality in human organizational relationships; and
selection and promotion of personnel solely on the basis of technical
competence. These ideas are institutions, visible in merit systems of
public employment, for example, and certainly in the IRB regulatory system.
Finally, Alvesson and Deetz (1996) suggest “Taylor’s and Weber’s treatment
of rationalization and bureaucratization showed from the start the
corporation as a site of the development of modernist logic and
instrumental reasoning” (p. 194). The IRB system creates a thorough
illusion that Weber, Fayol, and Taylor were published yesterday.
Structural/Functional explanations. In the 1950s,
structural-functional theory dominated sociology, the emphasis placed on
the power of institutionalized norms to determine behavior. Beginning in
the 1960s, sociologists began to abandon this thinking. In 1967, Garfinkel
rebelled, arguing the deterministic models presented people as “judgmental
dopes” who couldn’t do their own thinking. It’s fairly clear that sometimes
we won’t do our own thinking. And other times, it appears, we cannot. Not
under the rules, anyway. The informed consent process in social sciences,
and the Theory X management techniques of institutional (and
institutionalized) IRBs have provided a system that treats both social
science researchers and participants as unthinking and imbecilic,
respectively (if not respectfully). In the IRB system, the process now
dictates much researcher behavior. But, as described in earlier chapters,
the IRB system is especially sensitive to the larger political context, and
therefore driven by various outside forces. Few of the changes produced are
directly related to human subjects protections, and even more rarely
related to the actual event that “started” a given debate (see p. 22, and
footnote # 172, p. 155). It is understandable why researchers might become
confused in making applications. Strategies that worked at one time (when
Ellis was director of OPRR vs. Koski as head of OHRP, for example, or rules
regarding stem cell research with Clinton as president vs. rules under the
Bush administration, or when the local IRB changes membership composition,
experiences outside criticism, or is sanctioned) are deemed (or “really”
become) ineffective (or less than optimally so) for no apparently good
reason, i.e., no evidence is provided to support a/the change(s)
proposed. Institutional regulatory responses to these outside (and some at
least arguably irrelevant) forces take a variety of forms, (most often,
rule or enforcement proliferation in response to a generalized need to “do
something,” see particularly the headlines in Brainard, 1999, Dec 17; 2000,
Feb 4; 2000, Apr 14; 2000, May 26a & b; 2000, Jun 2; 2001, Jan 5; 2001,
Jan 12; and 2001, Feb 16, as examples) and numerous activities occur
simultaneously (parties to the situation feel compelled to “do something,”
demonstrated in the same Brainard headlines cited above, as well as the
articles themselves), the synergy of which transcends individual events in
affecting the system.
de Saussure (1915/1966), a father of structuralism, stated that words
are not meaningful in isolation (consistent with G.H. Mead and others), but
receive their meanings because of their differences (i.e., in the minds
of the meaning makers, an anti people-are-dopes position) from other
words of the same systems. de Saussure used the term la langue
to describe the system of a language, which needs to be treated differently
than the actual speech of the people who use the language, which he
described as parole. La langue is a social product,
transmitted to the members of the linguistic community, and it comes to
exist "in the mind" of each speaker, affecting the way the world
is understood. I believe that even though the phrase “in the mind” is used,
what is being described is an example of (SINS) STRUCTURES,
INSTITUTIONALIZATIONS, NATURALIZATIONS, SIMULATIONS operating mostly out
of mind once they are learned. This lack of visibility contributes to
the power of these terms and processes. Many fewer questions arise once the
system is installed, once it is learned, whether it is “understood” or
“questioned” or not.
Entrenchment. The IRB regulatory system is so entrenched (full of
(SINS) STRUCTURES, INSTITUTIONALIZATIONS, NATURALIZATIONS, SIMULATIONS) it
no longer needs overt coercion to operate smoothly. Fear of litigation,
fear of sanctions, fear of personal humiliation, fear of financial loss are
coercion, and are often perpetrated (covertly) on oneself. These
bars (see p. 44) are seen as an inevitable part of research life.
IRBs have lots of carrots, especially as far as student researchers are
concerned, and unless IRB members are highly vigilant, they are inherently
coercive – students “know” they can’t publish (and will, therefore,
certainly perish attributable to other SINS beyond the scope of the present
study) without the approval of the IRB to live their research lives, or at
least finish their programs. IRBs control the definition of “timely” (for
example, what constitutes a timely response from the IRB or from
researchers, what constitutes a timely report to regulators about problems
with a study, and determining what is considered a “timely” submission or a
“timely” response from a local IRB. See also OU Policy, Section 6, parts 1
and 3; Section 11.2, part 1; Section 12.2, part 1) and are above student
scrutiny, i.e., the board doesn’t have to answer to student whying,
evident by the lack of due process or appeals structures. The OUIRB policy
regarding the use of human subjects states in Section 10, part 5, when the
OUIRB disapproves an investigator’s request{233}*,
“notification shall include a statement of the reasons for its decision and
the investigator shall be given an opportunity to respond in writing or in
person.” [The OUIRB] may, at its own discretion, “re-review and reconsider
its decision to disapprove a research activity at any time” (OU Policy,
Section 10, part 5). This could hardly be considered the intersection
of Adequate Appeal Avenue and Reasonable Regulation Road.
To provide guidelines, as the federal regulators do, suggesting what is right
is substantially less intrusive than requiring paperwork that is designed
(in futility and perpetuity at once) to ensure people do the right thing.
Rules do not replace a researcher’s personal values or lack of them, as
argued previously (see p. 27-29 and 244-246; see also Cassell, 1982). Trust
of researchers on the part of institutions allows “blanket assurance” from
researchers (and that is a leadership option they have available, i.e.,
no federal rules prevent it). But, a blanket assurance for researchers is
not “normal,” therefore, system staticity (i.e., inertia) being a
strong prevailing tendency, blanket assurance is less likely to be done at
the institutional level. If it were adopted, the focus of classroom
discussion might shift toward discussion about protection, because the
process would be minimal, simple, painless, mostly unremarkable and
forgettable, as, I would argue, a “good” process should be. There are, of
course, political and psychological, even financial benefits to making
things appear worse than they are,{234}*
namely the establishment of the illusion of the “need” to “do something” in
a given situation and that “doing something” would actually positively
impact anything (also see footnotes # 26 and 27, p. 22).
Under the proper conditions, such as clear, non-coercive written
agreements and the tendency to keep one’s word, employees (researchers)
seek out and accept responsibility and exercise self-control and
self-direction (McGregor, 1960). Policies like these are more likely to
develop a climate of trust. Distrust and coercion was demonstrated in the
past few years of federal regulatory activity, when more sanctions were
imposed in 20 months than in the 20 years prior (Brainard, 2000, Feb 4).
And, Congressional activities in 1997 (and those since) were attempts to
put a system promoting distrust and coercion in place, i.e., more
rules, more direct supervision, and (redundant) criminal punishment
provisions. These actions appear to have been attempts to enhance political
capital (see Joe Kennedy’s prepared statement to the U.S. Senate Committee
on Labor and Human Resources, U.S. Senate, 1994, Jan 13, p. 3-4). And,
Campbell (1999, May 28) quotes a university IRB official as saying “I think
it is more than just a coincidence that the Inspector General releases this
damning report about IRBs last summer (referring to DHHS OIG 1998b; see
also Campbell, 1998, Jun 12), Congress holds a hearing about it, then all
of a sudden OPRR comes out with this flurry of suspensions,” p. A30).
Diverse Voices Offer Similar Reasons for Change
The authors of the DHHS OIG reports and similar documents have concluded
that the IRB regulatory system is in need of reform, and diverse voices
have identified similar problems and offered similar reasons for and
suggestions about them. These voices include the following: OPRR, 1999;
NBAC, 2001; NIH, 1999; Monastersky, 2001, May 21; Greenberg, 2001, Jan 19;
Brainard, 2001, Jan 12; Brainard, 2001, Mar 9; Brainard, 2000, September
13; Wheeler, 1991, Dec 4 (who quoted long–time regulatory observer and Yale
professor of law, medicine, and psychiatry Jay Katz as having called
informed consent “a charade”); Burd, 1994, Feb 9; AAUP, 2001; ACHRE, 1995;
Brainard, 2000, Feb 4, Brainard, 2000, Mar 17, Brainard, 2000, May 30;
Brainard, 2000, Sep 13; Campbell, 1997, Sep 12; Campbell, 1998, Apr 3;
Charo, 1999, Jun 25; Charo, 1999, Mar 26; GAO, 1996; GAO, 2001; Geertz,
1988; Gray, 1982; Institutional review boards: A system in jeopardy,
1998, Jun 11; Hayes, Hayes & Dykstra, 1995; Healy, 1999, Jul 30; NBAC,
1997; OHRP, 2001, Jan 10; Okie, 2001, Aug 6; Pence, 2001, Jan 12;
President’s Commission for the Study of Ethical Problems in Medicine and
Biomedical and Behavioral Research, 1981; U.S. House, 2000, May 3; Punch,
1998; Tropp, 1982; and Wheeler, 1991, Dec 4.
“IRBs review too much, too quickly, and with too little expertise” (DHHS
OIG, 1998b, p. ii){235}*.
“This system was created to deal with a simpler research environment,"
says Arthur L. Caplan, director of the Center for Bioethics at the
University of Pennsylvania (as quoted in Walker, 1996, Nov 8, p. A29).
Caplan continues, “The volume of research proposals has gone up, the pace
of scientific activity has increased considerably" (p. A29). In
addition, some researchers and universities have acquired a financial stake
in the growing number of studies sponsored by private industry, Caplan
says, raising concerns about conflicts of interests. Some argue it no
longer makes sense to give university officials, whose institutions depend
on money brought in by government-sponsored clinical research, so much of
the responsibility for maintaining the ethical standards of research.
"The local character of these panels is more and more dubious,"
given the "amount of money at stake," says David J. Rothman,
professor of social medicine at Columbia University's College of Physicians
and Surgeons, and a member of that university's IRB (as quoted in Walker,
1996, Nov 8, p. A29).
This perceived need for reform (DHHS OIG 1998b; Brainard, 2000, Sep 13;
GAO, 2001, and others) is based on problems such as those found to be
inherent in bureaucracies, i.e., the system is slow moving and
reactionary, with a bias toward over-regulation, overly complex and
ambiguous federal regulations, and even more ambiguous institutional
renditions of them.{236}*
At the institutional level, (and based on the proliferation of rules
there), distrust of researchers and coercion appear prevalent. And as
mentioned, inconsistencies across IRBs exist, at least in part due to the
localized nature of the IRB system, and perhaps enhanced by the transience
of faculty/scholars. Finally, problems are apparent with new areas of
research that are not covered under the rules (as shown in the gene therapy
cases, particularly Gelsinger) and new/emerging/re-emerging unobtrusive
methods (many have argued unduly) restricted by the rules (see AAUP, 2001).
Several views, critical and others, are desirable in attempting to gain an
understanding of the system (an understanding that would be likely enhanced
by direct observation). That understanding must surely precede any
intelligent attempt to effectively (by any dogmatic standard) reform the
system. Compared to what it took to develop, maintain, expand, and entrench
the system, it appears it takes much more effort to reform (or eliminate or
ignore) it. And, to eliminate any parts of it will require, it appears,
substantially greater effort than the more frequent but still somewhat
rare{237}* “lateral revisions” (what we might call minor rule
changes that don’t “really” increase or decrease the burden of the
regulation nor change the effects of it in any significant way; i.e.,
change for the sake of change, activity for the sake of image). But to
convince people (researchers and regulators) that it can be eliminated
will, it is likely, take the most effort of all. (I recall a story about, I
believe, W.K. Kellogg. When asked how much money it takes to get into the
cornflakes business, he replied, “To build the facility? $X million. To get
people to buy them? Three times as much.”)
Even with the widespread voluntary adoption of regulations by
institutions (at least, supplying the text that “constitutes” adoption),
many research settings do not adhere to the Common Rule (45 C.F.R. § 46,
1991), intentionally or not. Some may interpret the Rule in any given
situation differently than regulators or lawmakers intended, or fail to
apply the Rule in instances when regulators would, and so on. Additionally,
there certainly must be some degree of intersubjective disagreement among
various IRB members, discontinuity with change in the board chair, outside
pressures, etc.
Still the larger questions are not related to how we have come to
accept, expect, dread being treated like incompetent, immoral idiots by
power-crazed jackasses{238}*,
but why ? Adorno (1989a, p. 133), states “conformity has replaced
consciousness” and observes advice that is valid every day is idiotic (see
Adorno quote below, p. 233).
Risks of Over-regulation
Charo (1999, Jun 25), professor of law and medical ethics at the
University of Wisconsin at Madison, and a member of the National Bioethics
Advisory Commission (NBAC) asks:
How many people have been used
in studies that lacked the basic protections of the Common Rule? Nobody
knows, because no law mandates the collection of data on human subjects
used in research. How many of those people thought they were patients
rather than research subjects? Nobody knows. How many were injured or
received substandard care? Nobody knows. How many of them paid, privately
or through their insurance, for the privilege of being unwitting subjects?
Nobody knows. (p. A64)
Given that criminal charges are rarely brought, even in medical trials,
and there is no explosion in civil damage suits, it appears that these
questions are slightly paranoid, especially as they might be applied to the
“minimal risk” conditions of much social scientific research “treatments,” i.e.,
involving participants who are not patients. While vigilance
is perhaps good, paranoia is debilitating. I would counter Charo’s
questions with (perhaps) paranoid ones of my own: How many studies aren’t
attempted because of stifling regulation? How many researchers have given
up attempts to study sensitive issues, unlawful activities, protected
classes, the important fringe? How many times? How many people might have
been helped if regulation hadn’t prevented certain studies?
Within the IRB system, it is not apparent that decisions are based on
widespread observation of the system. No data exist because (virtually) no
oversight exists.{239}*
How can we expect to understand a system such as the IRB if we can’t even
look at it closely to see how it operates?{240}*
IRB decisions about sanctions, “needed” rule changes, new (marketed as
better by regulators’ definitions) procedures, etc., are often based on
emergency responses to various pressures. These pressures are often
political ones, those induced and perpetuated by media and spin doctors.
This may contribute to over-regulation, i.e., over correcting after
an atrocity or compliance violation is discovered and becomes public (see
Koski’s comment in footnote # 178, p. 164, regarding hammers and
thumbtacks). And (process based) “compliance” has replaced (purpose based)
“protection” as the most salient set of issues—a simulation in which the
process obscures the purpose.
The prevalent belief that regulatory systems, including IRBs, cause
little if any “real” harm by over regulating is dangerous (see O’Connor,
1979, who compares research regulation with that of business, stating “The
record of federal attempts to regulate business is hardly encouraging as a
model to be followed,” p. 226). Too much regulation is the preferred
error when the health and safety of human subjects of research
(overwhelmingly patients) are involved, to be sure. But as indicated, that
situation is uncommon; punishment here has been generally for violations of
rules, not harm to people. But just because too much regulation may be
preferable if a true dilemma should exist, that does not render
over-regulation harmless (see O’Connor, 1979), and the harms are exponentially
higher when the problem is system-wide, and invisible. Further, news of
atrocities and abuses has increased the tendency toward more (over)
regulation. As mentioned, rules don’t work to give values to those who
don’t already have them, and lack of rules doesn’t take a person’s values
away. Yet, when a researcher defaults to what is perceived as a more doable
study (i.e., more doable under the burden of too many and too many
ill-fitting and/or irrelevant rules), this defaulting may be harmful. (See
my questions in answer to Charo, p. 231 above.)
Adorno (1989a) states “The attitudes which the culture industry calls
forth are anything but harmless. If an astrologer urges his readers to
drive carefully on a particular day, that certainly hurts no one; they will,
however, be harmed indeed by the stupefaction which lies in the claim that
advice which is valid every day, and which is therefore idiotic, needs the
approval of the stars” (p. 134).
The system as it presents itself at the OUIRB appears oftentimes to stifle
rather than facilitate the pursuit of knowledge (academic inquiry, at
least), but the hegemonic (positivistic) discourse also promotes
“thinking-in-a-box-ism” in a place where that kind of non-thinking should
be discouraged, rather than prescribed. Who can blame researchers,
especially students, for following a path of least resistance, especially
in the face of all the encouragement they receive in that regard? The
“real” pressure to finish one’s program of study, and tendencies to follow
the path of least resistance are anti-educational. Why should we expect
students indoctrinated in this way to become “educated” later? {241}*
It appears we “really” need to do something, and it seems apparent no
one else cares as much, has as big an incentive, (is in as much “danger”)
or is as qualified as the researchers themselves. Reform of the system
perhaps, then, begins with reforming our own attitudes about the system,
and about reform. And an acknowledgement of, no matter how confusing,
painful, or exhausting, the “real” nature of things.
|