Journal of Clinical Child and Adolescent Psychology Copyright # 2007 by

2007, Vol. 36, No. 2, 242–252 Lawrence Erlbaum Associates, Inc.

Offering Parents Individualized Feedback on the Results of Psychological

Testing Conducted for Research Purposes With Children: Ethical Issues
and Recommendations
Marie-josée Lefaivre
Department of Psychology, Dalhousie University

Christine T. Chambers
Departments of Pediatrics and Psychology, Dalhousie University and IWK Health
Centre, Halifax, Nova Scotia, Canada

Conrad V. Fernandez
Departments of Pediatrics and Bioethics, Dalhousie University and IWK Health
Centre, Halifax, Nova Scotia, Canada

Research protocols involving children often include psychological testing as part

of an assessment battery. Inclusion of such testing raises the question of whether
parents (or others) should be offered the individualized results of their chil-
dren’s psychological testing conducted for research purposes. The purpose of
this article is to provide a review of the ethical issues and principles associated
with individualized feedback of psychological testing conducted for research
purposes. Two hypothetical cases are offered to illustrate the complexities of
this topic. Detailed recommendations for the management of disclosure of the
results of psychological testing in research settings are also proposed.

Should researchers provide parents with individua- providing participants with aggregated research
lized results of their children’s psychological test- results (i.e., general summaries of study findings;
ing conducted for research purposes? Studies that Fernandez, Kodish, & Weijer, 2003; Partridge &
include psychological testing have the potential Winer, 2002). Several studies have demonstrated
to uncover important information that could that the practice of providing such feedback occurs
impact a child’s welfare, such as learning difficult- inconsistently across research disciplines, including
ies or symptoms of psychopathology. The topic of psychology (Fernandez, Kodish, Shurin, et al.,
providing individualized feedback on the results of 2003; Fernandez, Kodish, Taweel, Shurin, &
psychological testing in research can be viewed as Weijer, 2003; MacNeil & Fernandez, 2006). Recom-
somewhat comparable to the ethical issues raised mendations on how to handle the feedback of
in sharing incidental findings during neuroimaging aggregated results provide suggestions as to when
research (e.g., Illes et al., 2004) or sharing of com- participants should be offered the opportunity for
plex results such as genetic testing (e.g., Shalowitz results (i.e., during the informed consent process),
& Miller, 2005). However, to our knowledge, the type of information that needs to be included
nothing to date has been published in the psy- in the summary, and the timing of the availability
chology literature in which this ethical topic is of such summaries (i.e., after peer review of study
explicitly discussed, nor are there clear recommen- results; Fernandez, Kodish, & Weijer, 2003).
dations from professional bodies for how this type Although previous recommendations have
of feedback should be handled. focused on summaries of aggregated research
Recently, there has been consideration of the results, a further complicated issue lies in the offer
moral obligation and the ethical necessity of of providing results of participants’ individual
psychological testing included in a research proto-
col, irrespective of the study results. This issue war-
Correspondence should be sent to Marie-josée Lefaivre,
rants unique attention. Of note, the discussion of
Department of Psychology, Dalhousie University, 1355 Oxford
Street, Halifax, Nova Scotia, Canada, B3H 4J1. E-mail: feedback of individual research data is distinct from
[email protected] that which is legally mandated in circumstances of

242
 RECOMMENDATIONS FOR INDIVIDUALIZED FEEDBACK

clear harm to self or others or suspected cases of evaluate the psychometric properties of this new
abuse. It is also distinct from broader debriefing measure and its clinical utility in a hospital setting.
procedures in which participants are provided with Consequently, empirically validated measures of
additional details about the nature of an experi- depression and anxiety were also administered to
ment that involved some form of deception. provide an assessment of criterion validity. Should
The purpose of this article is to provide a review parents of adolescents who score in the clinical
of the ethical issues specific to individualized range for depression and anxiety on any of the
feedback. We do so by (a) presenting an overview measures be informed of individualized results?
of the available ethical standards and guidelines Does the fact that the adolescents are mature min-
promulgated by various professional bodies in ors influence the decision regarding to whom to
psychology, (b) relating ethical principles to the offer this type of feedback? How should adoles-
matter of providing individualized feedback to cents (or their parents) be told about the results,
parents, and (c) describing the potential impact and by whom? How does the decision to pro-
of these ethical issues on relevant components of vide—or not provide—such feedback impact
the research process. We also propose recommen- everyone involved in the research process (i.e.,
dations to facilitate the decision-making process1 youth, parents, researchers)?
in approaching the offer of a return of individua-
lized results to research participants.
Case 2
To illustrate and contrast how different
research situations may present distinctive ethical A university research laboratory is conducting a
issues around informing young participants and longitudinal study aimed at identifying psychoso-
their parents of psychological test results, we start cial determinants of school success in children
by providing two hypothetical research scenarios. between the ages of 4 and 7 years at risk for learn-
ing difficulties. Participants were recruited within
the community via public announcements and
Hypothetical Cases
through after-school programs. Participants were
asked to complete a short cognitive assessment
Case 1
battery administered by doctoral students under
A multidisciplinary team based in a pediatric the supervision of a clinical psychologist. The tests
teaching hospital is conducting a research study included in this assessment were well-established
examining psychological correlates of medical measures with sound psychometric properties.
complications in chronic disease management in The study methodology also included longitudinal
adolescents. Participants were recruited through follow-up with short-term (i.e., 3-month) and
the general pediatric clinic. Data collection for long-term (i.e., 2-year) reevaluation. Should par-
the study included chart reviews and a series of ents of the young participants be informed of
self-report questionnaires completed by parents individualized results on the cognitive assessment
and adolescents with the capacity to consent for battery? Could the provision of individualized
themselves (i.e., mature minors). The research results be construed as coercive enticement or fair
assistant was under the supervision of a licensed compensation for participation? What impact
psychologist who served as one of the study’s coin- could the timing of feedback have on the results
vestigators. This coinvestigator was not involved of subsequent follow-up assessments given the
in the clinical care of the mature minors. The ques- longitudinal design?
tionnaire package included a recently developed
depression and anxiety scale that aimed to assess
mood and anxiety symptoms in a pediatric popu- Available Guidelines for Providing
lation diagnosed with a chronic illness. A manu- Individualized Feedback to Parents
script describing the preliminary psychometric
properties of this scale was under peer review at Professional ethics codes addressing research
the time the study was being conducted. One of conduct in psychology have been established by
the study’s secondary objectives was to further the American Psychological Association (APA),
the Society for Research in Child Development
1
Researchers should consult their Institutional Review (SRCD) and the Canadian Psychological Associ-
Boards about the existence of possible guidelines for offering ation (CPA). These codes provide different empha-
individualized feedback, including institutional policies and ses and approaches when considering ethical issues.
government regulations around the access to information and
privacy legislation, but we hope our recommendations will also As a result, researchers are often encouraged to rely
assist Institutional Review Boards to develop policies around on more than one set of ethical guidelines when
this issue. planning and conducting research (APA, 2002;

243
 LEFAIVRE, CHAMBERS, FERNANDEZ

Ernst, 2003). We provide an overview of selected research purposes.2 The absence of detailed and
North American standards and guidelines (APA, comprehensive guidelines in terms of individualized
SRCD, and CPA) in terms of debriefing research feedback increases the probability of divergent and
participants in general and examine how these inconsistent interpretations of the ethical standards.
guidelines specifically relate to individualized feed- The SRCD code does go further by recognizing
back in the context of psychological research. the ethical responsibility of involving parents or
The two predominant ethics codes in the United guardians in research with children, which relates
States that guide psychological research with chil- to offering individualized feedback. ‘‘Principle 9:
dren are the Ethical Principles of Psychologists and Jeopardy’’ clearly states the following:
Code of Conduct of the APA (2002) and the Ethical When, in the course of research, information
Standards for Research with Children published by comes to the investigator’s attention that may jeo-
the SRCD (1990–1991). Both professional bodies pardize the child’s well-being, the investigator has
address the general matter of sharing information a responsibility to discuss the information with the
about the results of the study (vs. the specific parents or guardians and with those expert in the
matter of individualized feedback). ‘‘Section 8.08: field in order that they may arrange the necessary
Debriefing’’of the current APA code states assistance for the child. (p. xx)
This SRCD principle acknowledges parents’ and
(a) Psychologists [should] provide a prompt
guardians’ fundamental right and legal obligation to
opportunity for participants to obtain infor-
be actively involved in any decision that could poten-
mation about the nature, results, and conclu-
tially impact their children’s physical and psychological
sions of the research, and they [should] take
health, as well as their education (see also Bersoff &
reasonable steps to correct any misconceptions
Bersoff, 1999; Hesson, Bakal, & Dobson, 1993).
that participants may have of which the psy-
Unfortunately, this SRCD principle does not explicitly
chologists are aware.
operationalize what constitutes information jeopardiz-
(b) If scientific or humane values justify delaying
ing a child’s well-being, nor does it specify how feed-
or withholding this information, psychologists
back should be disseminated in such circumstances.
[should] take reasonable measures to reduce
Nonetheless, the various ethics codes seem to imply
the risk of harm.
that parents or, if appropriate, the mature minor
(c) When psychologists become aware that
should generally be made aware of any pertinent infor-
research procedures have harmed a partici-
mation that might impact the child’s safety and welfare.
pant, they [should] take reasonable steps to
minimize the harm. (p. 12)
The SRCD addressed sharing the results of a Ethical Principles in Providing Individualized
research project with participants in similar terms Feedback to Parents
under its ‘‘Principle 12: Informing Participants.’’
This principle states. In all decisions, researchers must strive to bal-
Immediately after the data are collected, the ance fundamental guiding principles such as the
investigator should clarify for the research parti- respect for human dignity, beneficence, and justice
cipants any misconceptions that may have arisen. that are the foundation of the codes’ standards.
The investigator also recognizes a duty to report Respect for a person’s dignity is one of the
general findings to participants in terms appropri- cardinal principles underlying ethical research.3
ate to their understanding. Where scientific or
2
humane values justify withholding information, The Canadian Code of Ethics for Psychologists (CPA, 2001)
offers similar principles and standards when discussing debrief-
every effort should be made so that withholding
ing research participants. The obligation of debriefing is mostly
the information has no damaging consequences focused within the context of incomplete disclosure. The con-
for the participant. cept of debriefing is only briefly mentioned without elaboration
These standards and principles particularly in terms of enhancing the knowledge of participants and pro-
focus on the obligation of debriefing research viding them with a sense of contribution. The CPA code does
not specifically address individualized feedback on results of
participants if there has been incomplete disclos-
psychological testing conducted within a research context.
ure or deception during any phase of the project. 3
The Canadian Code of Ethics for Psychologists (CPA, 2001)
They also recognize the importance of offering a goes further than any other code by attributing to it a predomi-
summary of the study’s results. Neither the APA nant importance compared to the other ethical principles. It
nor the SRCD ethical codes, however, provide states that, in absence of clear and imminent danger to the
explicit guidelines on the nature and extent of physical safety of an individual, ‘‘respect for the dignity of per-
sons’’ should be given more weight when faced with a dilemma
the information to be offered, nor do they stemming from conflicting principles. The CPA code of ethics
specifically address individualized feedback on strongly supports, however, that all of its guiding principles
results of psychological testing conducted for be considered and integrated in any ethical decision making.

244
 RECOMMENDATIONS FOR INDIVIDUALIZED FEEDBACK

Indeed, one of the strongest arguments in favor of avoided, but every research participant is entitled
offering individualized feedback to research parti- to fair compensation for their contribution to
cipants is that this procedure obligates researchers the study.
to treat each of their participants primarily as per-
sons or an end in themselves rather than as a
means to an end. Bersoff and Bersoff (1999) Ethical Issues Involved in Offering
pointed out that even if the individuals Individualized Feedback to Parents
conducting studies simply consider themselves
researchers, it does not preclude the participants Besides the different advantages, there are also
and their families to view these researchers as several viable arguments that suggest that indivi-
clinicians or people in a helping profession. dualized debriefing of psychological testing for
Indeed, Fisher (2003) conducted a study that research purposes may be associated with certain
examined adolescent and parents’ perspectives on risks. As a result, this section explores ethical
ethical issues around risky behavior (i.e., drug issues raised by the individualized return of results
use and suicide) research. She found that approxi- by discussing their potential impact on relevant
mately 40% of both groups believed that youth components of the research process. This section
who report drug problems on research question- highlights some of the clinical and scientific reper-
naires are asking for help. Approximately 70% cussions associated with these ethical issues.
of youth and parents adopted the same view when
asked about participants reporting suicidal ten-
Informed Consent Procedures
dencies. Researchers must not ignore the presence
of psychological distress in their young parti- The concept that parents have a right to
cipants simply because the data were collected relevant individual results of research should first
for research rather than clinical reasons (Bersoff be addressed before any assessment measures are
& Bersoff, 1999; Drotar et al., 2000). collected. The implications of offering feedback
In fact, ethical regulations also emphasize the are relevant in terms of recruitment strategies
importance of beneficence and the need to do no and the initial consent procedure. We believe that
harm (see ‘‘Principle A: Beneficence and Nonmale- the decision by the researcher to offer or not offer
ficence,’’ APA, 2002). Consequently, under the the individualized results of psychological testing
ethical principle of ‘‘beneficence’’ as well as needs to be clearly explained and justified to the
‘‘respect for persons,’’ researchers should provide participants during the consent procedure. The
information that may positively impact quality of availability of individualized feedback may influ-
life or reduce harm. Moreover, researchers are ence participants or their parents’ decision to take
obligated to provide any information to a parti- part in a study.
cipant’s parents or guardians to support the We also emphasize the offering of individua-
parents’ societal role in reducing harm and lized feedback rather than the automatic provision
enhancing the well-being of their child (Bersoff & of results, because informed participants and their
Bersoff, 1999; Hesson et al., 1993). guardians must also have the right to decline
Family advocacy groups have further argued such feedback. The decision to receive or decline
that the children and families who take part in stu- research results should be made only after the par-
dies are primary stakeholders, and they, conse- ticipant has been fully informed of inherent risks
quently, need to be more actively involved in the and benefits (Fernandez, Kodish, & Weijer, 2003;
full research process (Osher & Telesford, 1996). see also Standard 8.02 of APA, 2002). Parents
The procedure of offering feedback is likely to may prefer to not know the results of psychologi-
reduce the probability that participants may feel cal testing to avoid any labeling of their child or to
exploited or forgotten as key players in the avoid pathologizing what they may see as simply
research process (Fernandez, Kodish, & Weijer, eccentricities or unique personality characteristics.
2003). The opportunity to gain knowledge, includ- The emphasis on an offer is also in line with the
ing information provided in any type of debriefing APA’s 8.08 Debriefing Standard, which specifies
(e.g., aggregated results or feedback), is often cited that it is the opportunity to obtain the information
as one of the benefits of research participation that should be provided to participants.
(Eyde, 2000; Tri-Council Policy Statement, 1998 This ethical necessity of informing participants
with 2000, 2002 updates). Similarly, investigators about the availability or the absence of individua-
must also consider that participants and their fam- lized feedback during the consent procedure is
ilies are entitled to fairness and justice throughout not, however, without scientific consequences.
the research process (see ‘‘Principle D: Justice,’’ Stanton, Burker, and Kershaw (1991) experimen-
APA, 2002). Coercive incentives need to be tally manipulated consent forms to examine the

245
 LEFAIVRE, CHAMBERS, FERNANDEZ

impact of expected follow-up procedures on the to freedom to consent. Access to feedback on the
self-report of depressive symptoms in an under- individual results of psychological testing of one’s
graduate sample. They found that participants, child may represent undue pressure or incentives
especially men, who were told beforehand to expect that jeopardize a participant’s freedom to consent.
a more intrusive follow-up (e.g., experimenter may Researchers could diminish the possibility of
contact participant and a significant other based on coercion by putting in place measures such as
scores on a depression scale) were less likely to avoiding citing inducements in their recruitment
report depressive symptoms compared to parti- materials (e.g., study advertisements) and clearly
cipants who were informed of a less intrusive explaining, during the consent procedure, what
follow-up protocol (e.g., a list of community the individualized feedback will specifically consist
psychological resources provided to all participants of and its limitations.
and no further contact is offered). In another study The possible threat to freedom to consent as a
(Fisher, 2003), when asked whether waiving the result of coercive enticement also has research impli-
parental permission procedure or assuring confi- cations. As Attkisson, Rosenblatt, and Hoagwood
dentiality would have an important impact on the (1996) pointed out, the quality of data frequently
validity and honesty of youth’s responses on self- rests on the freedom of choice exerted by the parti-
report measures, a notable majority of the ado- cipants and their parents. The availability of indivi-
lescent responders answered in the affirmative. dualized feedback might bring about sample biases
These results generate important scientific and ethi- that could impact the generalizability of the study’s
cal concerns that place researchers in a potential results. Researchers who offer feedback as a recruit-
conflict of interest as to how they handle these ethi- ment strategy might be faced with parents and
cal concerns, because these decisions could subse- youths who falsely report difficulties or signs of psy-
quently affect the integrity of their data. chopathology simply to meet the eligibility criteria.
To guard against this possibility, researchers must
be sure to clearly define, a priori, inclusion and
Research Compensation and Coercion
exclusion criteria to minimize sample bias as a result
Some researchers raise concerns that the avail- of offering feedback.
ability of feedback as part of study protocols may
encourage potential participants to take part in
Multiple Roles and Obligations of Researchers
research to obtain assessment results that would
not otherwise be easily available. The main issue Another ethical conundrum in providing indivi-
around these practices stems from a hotly debated dualized results stems from the multiple relation-
and unresolved issue of what is considered excessive ships and obligations that researchers may hold.
enticement for research participation (Grady, In the context of offering individualized feedback
2005). Some view the availability of individualized to the parents of participants, this issue is parti-
feedback following participation in a study as an cularly relevant because researchers might find
appropriate way of compensating the participant’s themselves divided between respecting children’s
time commitment, whereas others are concerned right to privacy (linked to respect for autonomy)
about coercion, as access to psychological assess- versus parents’ moral and legal rights to advocate
ments are sometimes subject to significant waiting for their children’s welfare (Fisher, 2003; Hesson
lists in the public sector or considerable expense et al., 1993). The difference in a youth’s cognitive
to parents if obtained in private practice. Pro- ability and legal age to consent for a research
fessional bodies caution against coercive measures, study constitutes a complex and delicate reality
such as undue incentives, that seriously jeopardize wherein health professionals sometimes must
participants’ freedom to consent and, subsequently, balance their ethical standards with their legal
impacts the dignity of the participants. The SRCD’s obligations (see review in Hesson, et al., 1993).
(1990–1991) ‘‘Principle 5: Incentives’’ states. The ability to assent, and ultimately to fully
Incentives to participate in a research project consent to research, is recognized as a continuum
must be fair and must not unduly exceed the range unrelated to an arbitrary age. However, many sta-
of incentives that the child normally experiences. tutes, for the sake of simplicity, define a specific
Whatever incentives are used, the investigator age at which parental consent is required for
should always keep in mind that the greater the research, irrespective of the child’s cognitive
possible effects of the investigation on the child, maturity (Leikin, 1993; Wendler & Shah, 2003).
the greater is the obligation to protect the child’s In the context of research participation, most state
welfare and freedom. (p. x) laws recognize the autonomy of emancipated min-
The APA’s (2002) Standard 8.06 similarly cau- ors to consent to research; however, state-specific
tions against coercive practices, given the threat definitions of mature minors vary in whether this

246
 RECOMMENDATIONS FOR INDIVIDUALIZED FEEDBACK

recognition of autonomy also extends to this evidence for the clinical utility of the measures
group (Fisher, 2004). These variations make it dif- used in research protocols, when examining
ficult for the researcher to determine who may whether it is appropriate to offer individualized
have legal rights to individualized results (parents feedback.
vs. mature minors). Data with clinical ramifications are sometimes
collected to establish the psychometric properties
of a newly developed scale or assessment measure
Research Team Composition
(e.g., Case 1). In such a case, it is usually difficult
Research team composition must also be con- to justify offering individualized feedback to part-
sidered as teams and protocols are often very icipants before the reliability, validity, and clinical
diverse. Multidisciplinary team members may have utility of that measure have been established
different types of expertise, degrees of ethical obli- Drotar et al., 2000). In addition, research proto-
gation, and clinical and nonclinical professional cols are often insufficient in length and breadth
responsibilities. It is important to consider the type to make a clinical diagnosis (e.g., Case 2). For
of training and the experience level of the indivi- example, researchers often rely on cutoff scores
duals who are directly responsible for data collec- rather than diagnostic scores to divide participants
tion (Attkisson et al., 1996; Drotar et al., 2000). into groups or to conduct analyses. The scores
Furthermore, it is not only essential that qualified yielded by many psychological measures are more
professionals are available to competently super- closely aligned with a dimensional approach,
vise the administration, scoring, and interpretation which differs from the categorical diagnostic sys-
of the clinically relevant measures, but it is also tem of the Diagnostic and Statistical Manual of
necessary to have qualified professionals oversee Mental Disorders (fourth edition, text revision;
the sometimes sensitive process of offering indivi- American Psychiatric Association, 2000). In fact,
dualized feedback to participants (Bersoff & it is important to distinguish psychological testing
Bersoff, 1999; Eyde, 2000). The APA (2002) ethics from diagnostic assessments. The former involves
code addresses the issue of competence in Stan- simply the collection of data, whereas the latter
dards 2.01 and 9.07 (see also Standards 9.09 and is ideally a multimethod approach whereby
9.10). Therefore, the composition of a research psychological testing is only one of the various
team might have important ramifications on the tools employed to establish a diagnosis (Sattler,
design of the research protocol and the interpret- 2001).
ation and feedback of clinically relevant assess- Adding measures not described in the informed
ment measures and their dissemination. consent and not directly relevant to the research
Similarly, offering feedback to parents directly question to obtain a more detailed clinical evalu-
affects the researchers, at the very least in terms ation is ethically inappropriate and scientifically
of their time commitment, the availability of problematic. It risks, in part, changing the
resources, and financial impact (Bersoff & Bersoff, relationship of the researcher to one of a clini-
1999). As pointed out by Bersoff and Bersoff, a cian (Attkisson et al., 1996; Eyde, 2000). This is
researcher’s qualification level is an ethically irrel- particularly problematic in the case of studies in
evant issue when considering the basic rights of which the psychological assessments are incidental
research participants. Bersoff and Bersoff accu- to the research’s main purpose (Attkisson et al.,
rately stated that ‘‘all participants should have 1996; Eyde, 2000). Therefore, researchers risk hav-
the same basic rights simply by virtue of their ing a high false-positive or false-negative rate that
being in a study’’ and that ‘‘appropriate ethical could harm participants (Eyde, 2000).
safeguards should be determined by the nature of Given that the feedback provided to families is
the information being collected and not by what likely accompanied by several limitations due to
can be reasonably expected given the researcher’s the issues around collecting data beyond the
level of training and licensure’’ (p. 17). research protocol, it may be difficult to predict
how the young participants and their parents will
react when provided with results of psychological
Psychometric Properties and Clinical Utility
assessments. Principle 13 of the SRCD (1990–
of Assessments
1991) ethical standards acknowledges this reality
As illustrated in the case examples, measures by stating ‘‘because the investigator’s words
that provide individualized results have the poten- may carry unintended weight with parents and
tial to unmask clinically relevant signs of learning children, caution should be exercised in repor-
difficulties or symptoms of psychopathology. ting results, making evaluative statements, or giv-
Therefore, it is essential to consider carefully the ing advice’’ (p. xx). Consequently, it cannot
psychometric properties, as well as the available be overemphasized that the interpretation of

247
 LEFAIVRE, CHAMBERS, FERNANDEZ

psychological assessments is a very delicate and In some cases, it may be appropriate to wait until
complicated process that needs to be done with the end of a study before providing individualized
considerable tact (see also Standard 9.02b of feedback. In other cases, it may be necessary to
APA, 2002). Drotar et al. (2000) also cautioned provide more timely feedback due to the nature
researchers that parents might entertain unrealistic of the psychological issues being assessed or as a
expectations about the benefits their child will result of the duration of data collection.
experience by participating in a research study that The issue of coordinating services is important
offers feedback on psychological assessment. if participants are either currently receiving care
Finally, the provision of feedback to parents has from another health professional (e.g., pedia-
the potential of creating conflict between the par- trician) or likely to warrant some kind of pro-
ents and the child, especially if they disagree on fessional services (e.g., assessment from school
the best course of action. There is the risk that a psychologist). Although written in terms of thera-
child will be the subject of unnecessary labeling, peutic relationships rather than within a research
which could further impact his or her relationship context, Standard 3.09 (‘‘Cooperation with Other
with parents or encourage a self-fulfilling prophecy. Professionals’’) of the APA (2002) ethics code obli-
gates psychologists to ‘‘cooperate with other profes-
sionals in order to serve their clients=patients
Timing of Feedback and Follow-Up
effectively and appropriately’’ (p. 6; see also Stan-
Any individualized feedback suggesting a clini- dards II.18, II.19, and II.35 in CPA, 2001). Conse-
cally significant difficulty may suggest the need quently, it would seem unethical to deny the
for further assessment to clarify the nature of the parents individualized feedback on their child’s
problem. A researcher’s ethical and professional psychological assessment when it has already been
obligation to personally offer such assessment established that they will likely need such assessment.
versus the obligation to refer the participant needs The professional conducting the subsequent assess-
to be considered based on the ethical standards ment may be unaware that a child’s recent partici-
from different professional bodies. Although not pation in a research study with an assessment
addressed in the APA (2002) code, Standard II.31 component and administer the same measures.
of the CPA (2001) code of ethics stated the require- Alternatively, the clinician might be limited in
ment to ‘‘give reasonable assistance to secure the type of assessment that can be completed with
needed psychological services or activities, if per- the youth because of practice effects due to retest-
sonally unable to meet requests for needed psycho- ing. This situation is particularly relevant if such
logical services or activities’’ (p. 67). A researcher testing involves a cognitive assessment within a
may decide that he or she is unable to meet such 2-year period of research participation. Research-
a request given a conflict of interest based on mul- ers have to, therefore, carefully design methods
tiple relationships or on a lack of qualifications to to ensure that inclusion and exclusion criteria
deal with the particular problem. This reality could appropriately match their intended research popu-
be problematic in terms of the participant’s lation, including means to ascertain previous
welfare, if the referral implies that the youth psychological testing that may interfere with the
will be facing long waiting lists before obtaining scientific validity of their study and coordinate
the appropriate follow-up. One recommended with others who may be involved in similar testing
approach is to provide participants with a resource with the child in the future.
sheet with appropriate and timely referral options.
Furthermore, the timing of the feedback and
Recommendations for Providing
subsequent need for additional assessment could
Individualized Feedback to Parents
compromise the integrity of the research design,
especially in studies involving long-term follow-
Based on the consideration of the guiding
up. This, however, is not only accompanied by
principles of respect for persons, beneficence, and
scientific consequences but also by ethical con-
justice and existing professional psychology ethical
cerns, given that jeopardizing the validity of a
codes and standards, we propose the following
research design could in turn reduce the ethical
recommendations to investigators conducting
value of research participation (Rosenthal, 1994;
research with children that include psychological
cf. Parkinson, 1994). The nature of the assessment
measures:
results may also be time sensitive in terms of their
clinical utility. For example, the results of psycho-  Researchers should offer individualized feed-
educational assessments may become outdated back when well-validated psychological mea-
after 2 years of administration due to developmen- sures are administered, but only if these
tal or situational changes associated with the child. measures include either normative data or

248
 RECOMMENDATIONS FOR INDIVIDUALIZED FEEDBACK

empirically derived cutoffs (e.g., scores shown to  The results of the psychological testing should
predict clinical diagnosis or risk). Researchers be presented similarly to a brief psychological
should not provide feedback for newly deve- report using lay vocabulary that can be easily
loped and unvalidated measures, and measures comprehended by parents. The performance of
without sound psychometric properties for the participants should be described relative to
which no comparative data are available. established norms or cutoff scores. The inclusion
 Details about the availability of feedback must be of percentile ranks (provided that a clear expla-
clearly explained to the child and the parent during nation of percentiles is included) or a general
the consent procedure and reinforced as the study description of whether the result falls in the aver-
concludes. These details ought to explicitly outline age or below or above average range might also
what specific information will be included and what be appropriate. An example of such a descrip-
type of information will not be provided as well as a tion is provided in the appendix.
reason for this plan. The consent procedure also  The diagnostic limits of the feedback and the
needs to specify who will be involved in providing psychometric properties of the measures for
the feedback, who will access the feedback, and the particular population in lay language must
when and how the feedback will be provided. be clearly specified in the written feedback.
 Young participants, especially mature minors,  Psychological researchers do not have a personal
should be actively involved in (and aware of) their responsibility to provide clinical follow-up with
parents’ decision as to whether to obtain feedback the participants, unless there is a risk of harm
on their psychological testing during research. to self or others (e.g., suicide risk). They do have
The choice of a youth with the capacity to consent an obligation to provide appropriate and timely
should be valued and respected. Conversely, par- referral options to participants.
ents and researchers also need to be aware that  Researchers are encouraged to highlight the
younger participants might not have attained the need and importance of informing any person
developmental and cognitive maturity needed to conducting further psychological testing with
comprehend the complexity involved in being the child about the feedback information around
provided with individualized feedback. the measures administered during the study. The
 Feedback needs to be delivered in a manner that researcher should provide access to lists of com-
is as sensitive and considerate as possible to pro- pleted measures and results to clinicians in a
tect participants and their families’ dignity and timely manner.
welfare. The appropriateness of a particular type
of feedback is likely contingent on the popu-
lation studied, the research design, and antici- The Two Cases Revisited
pated consequences of the results to individuals
or their families. Written feedback in lay lan- From the previous discussion, it is clear that
guage should always be provided. In addition, offering individualized feedback on the results of
face-to-face feedback by a qualified research psychological testing to young participants and
team member is important when there are highly their parents represents a complex decision involv-
complex or ambiguous results, or when there are ing numerous ethical, clinical, and scientific impli-
significant negative implications to the individ- cations. We now revisit the hypothetical cases in
ual participant (e.g., clinical diagnosis). light of the previous discussion to illustrate and
 Given the complexity and sensitivity associated contrast how these research situations emphasize
with the interpretation of psychological assess- distinct ethical issues.
ments, the process of offering feedback should be
supervised by a professional qualified to provide
Case 1
feedback on psychological data. The level of
involvement of this professional would depend on This case illustrates the multiple roles and
the nature of the feedback provided to participants. obligations of researchers with respect to
 Specific information provided during feedback individualized feedback, as well as issues around
must include the title and purpose of the study psychometric properties and clinical implications
as they appear on the informed consent form of measures. The sample in this case is composed
to minimize revealing overly sensitive infor- of adolescents who are mature minors. As a result,
mation. The feedback must also include the date the research team is faced with a dilemma in which
of participation, the contact information of the they have to consider respecting the young parti-
principal investigator, the name and type of cipants’ right to privacy as well as their parents’
assessment measures, and who administered moral and legal rights to be made aware of any
the measures. information related to their children’s welfare.

249
 LEFAIVRE, CHAMBERS, FERNANDEZ

This situation might be particularly delicate if the lation norms might not specifically include norms
adolescents and their parents disagree on whether for youths with chronic illnesses. Nonetheless, it
they want to be offered the results of the psycho- might still be appropriate to provide the results
logical measures.4 of the well-validated measures as long as the clini-
The impact of this ethical issue might be mini- cal psychologist supervising the provision of the
mized by clearly explaining the details and the type individualized feedback specifies the interpretative
of feedback offered to the youths and their parents limits of the measures following the guidelines
during the consent procedure rather than waiting offered in the questionnaires’ manuals.
until after the data have been collected. It also pro-
vides the opportunity for adolescents and their
Case 2
parents to ask questions or obtain clarifications
about the scope of the feedback before consenting The ethical concerns around research compen-
to the study. sation and coercion, as well as the issues associated
If participants or parents agree to receive indivi- with the timing of feedback and follow-up, are the
dualized feedback, offering the youths a choice of main themes illustrated in this hypothetical case.
follow-up options might further minimize conflict. In contrast to Case 1, the participants in this case
These options should be based on our suggested represent a nonclinical population recruited from
recommendations for offering individualized feed- the community and may, therefore, have more lim-
back. This possibility for follow-up options can be ited contact with the health care system. There is,
more easily considered in this case, given that the however, a high probability that these participants
participants were recruited through their pediatri- will require some type of educational assessment
cian’s clinic. These follow-up options could for clinical purposes in the near future if they
include, for example, choosing to first inform their are, indeed, at high risk for learning disabilities.
parents or their pediatrician about the results of As a comprehensive assessment can be subject to
their psychological testing. This suggestion might waiting lists or considerable expenses, the avail-
also prevent follow-up procedures that are per- ability of individualized feedback provided free-
ceived as overly intrusive by the participants and of-charge might be appealing for families.
that, in turn, could increase the probability that To minimize ethical concerns around research
they would under- or overreport symptoms compensation and coercion, researchers conducting
(Stanton et al., 1991). In this case, the option of this study should explicitly clarify, during the con-
informing the participant’s pediatrician might also sent procedure, the details and the type of feedback
facilitate the coordination of health services and that will be offered to parents, including the type of
the accessibility to improved clinical care. information that will not be provided. This practice
This hypothetical case also illustrates issues would likely diminish the chances that the families
around the psychometric properties and clinical will entertain false expectations about the benefits
utility of the measures included in the protocol. of participating in the study. Moreover, researchers
On one hand, the newly developed measure of should not explicitly advertise the availability of
depression and anxiety specifically aims to assess such feedback as their primary recruitment strategy.
mood and anxiety symptoms in a pediatric popu- By not emphasizing the opportunity for feedback in
lation diagnosed with a chronic illness. The infor- the recruitment announcements (e.g., community
mation gathered by this measure might be posters), researchers would help minimize the pres-
particularly useful to this target population. The ence of undue pressure or incentives. This procedure
measure itself, however, is still under peer review, might also help diminish the introduction of poss-
and its psychometric properties are not yet clearly ible sample biases created by parents who may inac-
established, which makes them unsuitable as the curately answer questions around the inclusion or
basis of individualized feedback at this time. On exclusion criteria to access the individualized feed-
the other hand, the empirically validated measures back. The researchers need to carefully define
of depression and anxiety have sound psycho- inclusion criteria to avoid sample biases based on
metric properties and clinical utility, but the popu- recruitment issues. Due to practice effects following
repeated administration of measures, researchers
4
Although a full discussion on the matter goes beyond the must further consider carefully defining unambigu-
scope of this article, the clinical implications of the data col- ous eligibility criteria to assess and diminish the risk
lected might also place the researchers with a legal obligation of practice effects with participants.
to follow-up with participants answering questionnaire items As the participants in this study might be at
indicating that they are at risk of harm (e.g., suicidal risk, child
abuse). This particular type of follow-up might necessitate higher risk of requiring an educational assessment
breaking confidentiality and following legally mandated report- in the near future, it will also be essential that
ing procedures. researchers highlight to parents the importance of

250
 RECOMMENDATIONS FOR INDIVIDUALIZED FEEDBACK

sharing with any person conducting further psycho- In K. Hoagwood, P. S. Jenser, & C. B. Fisher (Eds.),
logical testing the information around the pre- Ethical issues in mental health research with children and
adolescents (pp. 47–57). Mahwah, NJ: Lawrence Erlbaum
viously administered measures, regardless of Associates, Inc.
whether the parents agree to obtain individualized Bersoff, D. M. & Bersoff, D. N. (1999). Ethical issues in the col-
feedback. If families opted for obtaining individua- lection of self-report data. In A. Stone, J. Turkkan, C.
lized feedback, researchers would be encouraged to Bachrach, V. Cain, J. Jobe, & H. Kurtzman (Eds.), The
provide such feedback by following our recommen- science of self report: Implications for research and practice
(pp. 9–24). Mahwah, NJ: Lawrence Erlbaum Associates,
dations detailed previously. In the alternative, the Inc.
researchers should still provide families that decline Canadian Psychological Association. (2001). Companion man-
such feedback an abbreviated feedback form simply ual to the Canadian Code of Ethics for Psychologists
stating the title and purpose of the study, the date of (3rd ed.). Ottawa, Ontario, Canada: Author.
participation, the contact information of the princi- Drotar, D., Overhosler, J. C., Levi, R., Walders, N., Robinson,
J. R., Palermo, T. M., et al. (2000). Ethical issues in con-
pal investigator, the name and type of assessment ducting research with pediatric and clinical child popula-
measures with clinical utility, and the name of the tions in applied settings. In D. Drotar (Ed.), Handbook
person(s) who administered the measures. This is of research in pediatric and clinical child psychology
especially relevant in Case 2 because it does not (pp. 305–326). New York: Kluwer-Plenum.
explicitly state whether school psychologists or Ernst, K. (2003). Textbook coverage of ethical considerations
in research with children. Teaching in Psychology, 30,
health professionals are following the participants. 253–256.
Finally, the consent around feedback should be Eyde, L. D. (2000). Other responsibilities to participants. In
revisited with parents at the two follow-ups. This B. D. Sales & S. Folkman (Eds.), Ethics in research
way, parents would be given the opportunity to with human participants (pp. 61–73). Washington, DC:
revise their choice as to whether they want to be pro- American Psychological Association.
Fernandez, C. V., Kodish, E., Shurin, S., & Weijer, C. (2003).
vided with the feedback. Offering to return results to research participants: Atti-
tudes and needs of principal investigators in the children’s
oncology group. Journal of Pediatric Hematology=
Oncology, 25, 704–708.
Conclusions
Fernandez, C. V., Kodish, E., Taweel, S., Shurin, S., &
Weijer, C. (2003). Disclosure of the right of research part-
Respecting the participant as a person is a fun- icipants to receive research results: An analysis of consent
damental responsibility of researchers. It is often forms in the Children’s Oncology Group. Cancer, 97,
possible to offer individualized results of psycho- 2904–2909.
Fernandez, C. V., Kodish, E., & Weijer, C. (2003). Informing
logical testing to research participants, but there
study participants of research results: An ethical impera-
are many caveats that must be considered, includ- tive. IRB: Ethics and Human Research, 25, 12–19.
ing reliability and validity of the data, minimizing Fisher, C. B. (2003). Adolescent and parent perspectives on
coercion to participate, conflicted or unclear roles ethical issues in youth drug use and suicide research. Ethics
of clinician and researcher, and potential harms and Behavior, 13, 303–332.
Fisher, C. B. (2004). Informed consent and clinical research
such as labeling. There is a strong need for empiri-
involving children and adolescents: Implications of the
cal studies investigating the ethical and scientific revised APA ethics code and HIPAA. Journal of Clinical
implications of providing parents with individua- Child and Adolescent Psychology, 33, 832–839.
lized feedback of results of psychological testing Grady, C. (2005). Payment of clinical research subjects. Journal
conducted for research purposes with children. of Clinical Investigation, 115, 1681–1687.
Hesson, K., Bakal, D., & Dobson, K. S. (1993). Legal and ethi-
Such studies could include an examination of psy-
cal issues concerning children’s rights of consent. Canadian
chologists’ current practices in providing feedback, Psychology, 34, 317–328.
the effects of feedback on parents’ attitudes Illes, J., Kirschen, M. P., Karetsky, K., Kelly, M., Saha, A.,
toward research, and child outcomes as a result Desmond, J. E., et al. (2004). Discovery and disclosure
of the feedback. of incidental findings in neuroimaging research. Journal
of Magnetic Resonance Imaging, 20, 743–747.
Leikin, S. (1993). Minors’ assent, consent, or dissent to medical
research. IRB: Human and Research Ethics, 15, 1–7.
References MacNeil, S. D. & Fernandez, C. V. (2006). Informing research
participants of research results: Analysis of Canadian
American Psychiatric Association. (2000). Diagnostic and stat- university-based research ethics board chairs. Journal of
istical manual of mental disorders (4th ed., text rev.). Medical Ethics, 32, 49–54.
Washington, DC: Author. Osher, T. W. & Telesford, M. (1996). Involving families to
American Psychological Association. (2002). Ethical principles improve research. In K. Hoagwood, P. S. Jenser, & C. B.
of psychologists and code of conduct. Retrieved January Fisher (Eds.), Ethical issues in mental health research
16, 2005, from http://www.apa.org/ethics. with children and adolescents (pp. 29–39). Mahwah, NJ:
Attkisson, C. C., Rosenblatt, A., & Hoagwood, K. (1996). Lawrence Erlbaum Associates, Inc.
Research ethics and human subject’s protection in child Parkinson, S. (1994). Scientific or ethical quality? Psychological
mental health services research and community studies. Science, 5, 137–138.

251
 LEFAIVRE, CHAMBERS, FERNANDEZ

Partridge, A. H. & Winer, E. P. (2002). Informing clinical trial this score is average (16th to 84th percentiles),
participants about study results. Journal of the American above ( > 84th percentile) or below ( < 16th per-
Medical Association, 288, 363–365.
Rosenthal, R. (1994). Science and ethics in conducting, analyz-
centile) average.
ing, and reporting psychological research. Psychological The Peabody Picture Vocabulary Test–Third
Science, 5, 127–134. Edition (PPVT–III) is a test of receptive (hearing)
Sattler, J. M. (2001). Assessment of children: Cognitive applica- vocabulary. Receptive vocabulary is your child’s
tions (4th ed.). La Mesa, CA: Author. ability to understand words. On this test, your
Shalowitz, D. I. & Miller, F. G. (2005). Disclosing individual
results of clinical research: Implications of respect for
child received a percentile score of [#]. This is con-
participants. Journal of the American Medical Association, sidered [below average=average=above average].
294, 737–740. The Expressive Vocabulary Test (EVT) is a test
Society for Research in Child Development. (1990–1991). Ethical of expressive (speaking) vocabulary. Expressive
standards for research with children. Retrieved January 16, vocabulary is your child’s ability to say words that
2005, from http:==www.srcd.org=ethicalstandards.html.
Stanton, A. L., Burker, E. J., & Kershaw, D. (1991). Effects of
are meaningful to others. On this test, your child
researcher follow-up of distressed subjects: Trade off received a percentile score of [#]. This is con-
between validity and ethical responsibility? Ethics and sidered [below average=average=above average].
Behavior, 1, 105–112. Please note that these vocabulary tests were
Tri-Council Policy Statement: Ethical Conduct for Research given as part of a research study by a trained
Involving Humans (1998, with 2000, 2002 updates). Retrieved
January 16, 2005, from http:==www.pre.ethics.gc.ca=english=
research assistant who was supervised by a
policystatement=policystatement.cfm. licensed psychologist. This does not take the place
Wendler, D. & Shah, S. (2003). Should children decide whether of a proper language or educational assessment. A
they are enrolled in nonbeneficial research? American below average score does not necessarily mean
Journal of Bioethics, 3, 1–7. your child has a learning or language problem. If
you have any concerns about your child’s learning
or language skills, you should discuss these with
Appendix your family doctor or pediatrician. They may
arrange for further testing.
Example of Feedback Letter Should your child need a speech and language
[Date] or school assessment within the next 2 years, it is
important that the tester knows that your child
[Address] has undergone these tests as part of a research
study.
Dear [parent’s name]: Should you have any questions about this feed-
back, please contact [name of principal investigator]
Thank you for participating in our recent study at [phone number]. Thank you again for helping us
at the [name of site] titled ‘‘[. . .]’’. The purpose of with this valuable research study!
this study was [. . .]. You asked for feedback as
to how your child did on the two vocabulary tests Sincerely,
that were given [date of participation]. This letter
summarizes how your child [child’s name] did rela- [Name of principal investigator
tive to other boys and girls his=her age.
Scores are given as a percentile. A percentile Affiliation and contact information]
means the number of children who scored below
your child out of 100 children of the same age.
For example, the 40th percentile means that your
child performed better than 40% (40 out of 100) Received May 8, 2006
children his=her age. We will then tell you whether Accepted November 7, 2006

252