Professional Documents
Culture Documents
Ethical Challenges
Ethical Challenges
doi:10.1093/inthealth/ihw020
ORIGINAL ARTICLE
Ethical challenges in research with orphans and vulnerable children:
a qualitative study of researcher experiences
Maureen C. Kelleya,b,*, Tracy Brazgb,c, Benjamin S. Wilfonda,b, Liliana J. Lenguad, Beth E. Rivine,
Susanne P. Martin-Herzf and Douglas S. Diekemaa,b
a
Department of Pediatrics, University of Washington School of Medicine; bTreuman Katz Center for Pediatric Bioethics, Seattle Children’s
*Corresponding author: Present address: The Ethox Centre, Nuffield Department of Population Health, University of Oxford,
Old Road Campus – Roosevelt Drive, Oxford OX3 7LF; Tel: +44 186 561 7805; E-mail: maureen.kelley@ethox.ox.ac.uk
Background: Orphans and vulnerable children (OVCs) represent a significant population worldwide, enduring
poor health and living conditions. Evidence-based interventions are needed. However, without parents, ethical
concerns about including OVCs in research persist. The aim of our study was to better understand the ethical
challenges facing researchers who work with OVCs.
Methods: We conducted semi-structured interviews with 12 international pediatric researchers working with
OVCs in seven countries. We used descriptive content analysis to characterize the ethical rationale for inclusion
and associated challenges.
Results: Researchers believed research was justified as a necessary means for informing evidence-based inter-
ventions to benefit OVCs directly or as a population. Ethical challenges included difficulty identifying OVCs given
variation among children living without parents; difficulty identifying guardians among a range of caregivers;
concerns about meaningfulness of guardian consent; difficulty assessing risk; and responding to children’s
many needs.
Conclusions: A range of caregivers bear responsibility to protect OVC’s interests in place of parents in research but
are often not prepared to do so. This places greater burden on researchers to assess risks and respond to
children’s needs. Findings suggest that we should improve support and rethink the roles of guardians, researchers
and older children in research participation and protection.
Keywords: Assent, Consent, International research ethics, Orphans, Vulnerable children, Vulnerable populations
have lost or been separated from parents following war and pol-
Background
itical conflicts in Afghanistan, Iraq and Syria.2 That we see such
Orphaned, homeless and other unaccompanied children and varied estimates of unaccompanied children and young people
young people—often referred to internationally as orphans and reflects the challenges of mapping and reaching a still largely in-
vulnerable children, or OVCs—represent a significant but poorly visible and transient population.
represented global population. UNICEF estimates that there are The health and social impact of living on one’s own as a child or
approximately 153 million orphans worldwide.1,2 Sub-Saharan teen is significant. These children endure dangerous living condi-
Africa and Southeast Asia continue to have the highest number tions, sexual exploitation, violence and exposure to the elements.
of orphaned children due to parental loss from HIV/AIDS, with Equally devastating is the under-reported impact of undiagnosed
17.9 million children estimated to have lost one or both parents and untreated childhood illnesses: high rates of preventable infec-
to the disease.1,2 In high-income countries, the numbers of tions, lack of access to vaccinations and dental care, treatment
unaccompanied children living without parents are still significant and prevention of sexually transmitted diseases, malnutrition, and
relative to available resources and child welfare institutions. For the long-term psychological impact of stress.4–8 Innovative, effect-
example, the US government estimates that approximately ive solutions to address the social, economic and health conditions
380 000 children are living without families—that is, living in tran- in which these children live and struggle are needed. There are many
sient residence, on the street, or in transitional institutions.3 Thou- well-meaning humanitarian and governmental organizations,
sands of unaccompanied children and young people in Europe large and small, offering aid to vulnerable children; however, the
© The Author 2016. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com.
187
M. C. Kelley et al.
188
International Health
Database No. studies that appear to No. studies conducted in the No. studies conducted in
enroll OVCs United States (% of N) international settings (% of N)
Note: These do not include general pediatric clinical trials that might enroll orphans in addition to children with parents, since this number is
not tracked.
189
M. C. Kelley et al.
for permission to schedule a telephone interview. Upon agree- I think they are a vulnerable population for a lot of reasons, and
ment, participants were contacted by follow-up e-mail to I think, as you would say for any particularly vulnerable popu-
arrange a time for the interview. Before the interview portion of lation it isn’t probably ethical to conduct high risk research on a
our study, all participants were given pre-interview demographic vulnerable population, if it could be just as well conducted on
questions to complete, and a consent form to be discussed at another population. (Participant 5)
the time of the interview. Oral consent was obtained at the begin-
ning of each audio-recorded telephone interview. Detailed partici- Another participant referenced the history of exploitative experi-
pant demographics and mention of specific study details have ments with orphans to underscore what he felt to be the
been omitted to protect researcher confidentiality. central ethical worry, which is that orphans had historically
been used as a convenience sample:
I think what people were trying to get away from is the fact
At the same time, these researchers were experienced with the However, when asked how they weigh the background, daily risks
daily life and vulnerabilities of children living without parents. To and stress in the lives of OVCs, researchers were hesitant to enroll
gauge researchers’ attitudes toward acceptable risk thresholds OVCs in research if they thought doing so might cause added
in research, we presented brief descriptions of different types of stress or trauma. Several researchers felt this should be true for
pediatric research attended with greater levels of risk. All but any children in research, particularly when working in low resource
one participant agreed that research involving greater than settings.
minimal risk that does not offer the prospect of a direct benefit
to the child being enrolled was not ethically justifiable for inclusion If you had a child who had, maybe, a serious mental or phys-
of OVCs. When probed, several argued that the benefit to all chil- ical ailment that was not the focus of the study, that the study
dren, or even the population of OVCs, would not warrant using one itself could be harmful to the child, or the information from the
very vulnerable child as a means to that end. Others offered more child in that study would not really be useful because of the
nuanced accounts that considered the severity of the disease situation of the child, particularly on disability grounds. … we
burden on children, slightly greater than minimal risk, and the would never want to continue trying to recruit a child if there
availability of non-orphaned children to include in research. was any harm that could come to the child. (Participant 1)
190
International Health
Several researchers noted personal challenges in navigating Variation was especially pronounced in contexts where informal
ethics approval for research involving vulnerable children and custody arrangements are the norm, and where children who
worried that barriers to approval might create a disincentive for would technically be considered orphans (have lost both
including OVCs in potentially beneficial research. parents) are not viewed as orphans as long as they had relatives,
neighbors or community members to care for them.
I think my biggest worry is that people won’t do this kind of
research because they’re fearful and they don’t have the If a child loses their parents, they’ve still got parents in the
knowledge to be able to know what can and can’t be done. form of uncles or the grandparents. Our notion of orphans,
… I think doing research [in my case] with homeless kids— in social terms, doesn’t really apply in many parts of Africa.
you can count the number of people who do intervention re- (Participant 3)
search on one hand. That’s a big problem! And I think it’s
Investigators conducting research in sub-Saharan Africa noted
because many people are fearful because they don’t under-
191
M. C. Kelley et al.
whose parents are both alive. It’s probably the most common children. For example, this participant was responding to a study
scenario here for children, is that they are living with a relative vignette describing a minimal risk observational study of play be-
and their parents are migrant laborers. So, it’s very, very havior in a school or orphanage.
common for children not to be living with a legal guardian,
I think it’s a question of study aim. Clearly, I could imagine
or a parent even when their parents are alive and working
studies in which asking these kinds of questions about
and everything, so it definitely took a little bit of thinking and
orphans, particularly, would provide us with this research
talking through with the ethics board, here. (Participant 5)
that would be of benefit to orphans and vulnerable children,
Those working in low-income settings also reported that several themselves. In which case I do think it’s ok to enroll them.
adult caregivers might be considered responsible for an orphaned Especially because I could imagine that play behavior would
child, taking turns bringing a child to the research clinic. Whichever be something that would potentially be affected by orphan
adult presented with a child was typically assumed to be the status. That would be something that would deserve study,
192
International Health
structures for OVCs. Particularly for researchers working with chil- or school supplies. Researchers did not report concerns about
dren living on the streets, or in regions with high rates of child- such compensation being coercive as benefits were addressing
headed households, it was often unlikely or difficult to locate an basic needs and the research was viewed as beneficial. Their con-
adult caregiver responsible for the child. cerns with compensation primarily pertained to the broader com-
munity, or other children not enrolled in the study. Several
It’s important to make sure that the inclusion procedures
researchers mentioned the challenges of offering even basic ben-
include, especially informed consent, have sufficient flexibility
efits to children in the study when other children, and other
within them, to make sure that the needs of children who have
members of the community, were also hungry or in need of
been orphaned can be recognized. [S]ome of the children may
basic supplies for the home or school, echoing familiar challenges
well have had guardians who were able to answer for them
in offering fair benefits in a resource-poor community. Research-
and to indicate their consent. Other children, the individual
ers described how well-meaning attempts to meet those needs
to whom informed consent ought to be directed was not
193
M. C. Kelley et al.
I make sure that they’re already tied into resources, and I feel Our findings do raise questions about prioritizing the role of guar-
like that’s kind of an ethical obligation. So, that they’re getting dians in consent and ongoing support of vulnerable children
the help that I think that they need and that they’re willing to enrolled in research. Results suggest substantial weight is being
accept. […]You know, for instance, if we are talking about some placed on the responsibility of guardians without reflecting the lim-
pretty deep depression, even if it’s not active suicide ideations, itations and variations of that role. Researchers in our study reported
of working with them, getting their consent first, but actually that any adult available is providing consent, and most did not view
taking them, walking them to somebody who can help with evaluation of a guardian as practical. While researchers reported
that. (Participant 12) many very caring and attentive guardian caregivers, they also
raised significant concerns about the real constraints on the head
Researchers reported resourceful ways of responding to children’s
of a busy orphanage, or new foster parent, or temporary caregiver
basic needs, often at personal expense, but most were done
to know a child well and advocate in the way parents do in the clin-
without institutional or sponsor support. Researchers recognized
194
International Health
were secondary to responding to immediate needs. As men- research are blurred and the need for trusting adult relationships
tioned, researchers expressed a strong sense of personal respon- is so pressing.
sibility to respond to the children’s basic needs and were aware
of potential conflicts of interest but did not discuss clear solu-
tions to the question of balancing these considerations. Nor did Conclusion
they speak to the broader moral expectations of the community A range of caregivers bear responsibility to protect OVC’s interests
for researchers to aid and ‘take in’ orphans, as others have in research in place of parents but are often not prepared to do so.
reported.24,25 This finding is consistent with other studies involv- This places greater burden on researchers to assess risks and
ing fieldworkers’ sense of personal obligation and lack of system- respond to children’s needs. Our findings suggest that we
atic processes for managing the often overwhelming needs of improve support and possibly rethink the roles of guardians,
participants.24–26,37,38 researchers, and older children in research participation and pro-
195
M. C. Kelley et al.
4 American Academy of Pediatrics. Developmental issues for young 26 Kingori P. Experiencing everyday ethics in context: frontline data
children in foster care. Pediatrics 2000;106:1145–50. collectors perspectives and practices of bioethics. Soc Science Med
5 Frank DA, Klass PE, Earls F, Eisenberg L. Infants and young children in 2013;98:361–70.
orphanages. Pediatrics 1996;97:569–78. 27 Attride-Stirling J. Thematic networks: an analytic tool for qualitative
6 DiGiuseppe DL, Christakis DA. Continuity of care for children in foster research. Qual Res 2001;1:385–405.
care. Pediatrics 2003;111:e208–13. 28 Hsieh HF, Shannon SE. Three approaches to qualitative content
7 DuPlessis HM, Cora-Bramble D. Providing care for immigrant, homeless, analysis. Qual Health Res 2005;15:1277–88.
and migrant children. Pediatrics 2005;115:1095–100. 29 Millum J, Emmanuel EJ. The ethics of international research with
8 Ensign J, Panke A. Barriers and bridges to care: voices of homeless abandoned children. Science 2007;318:1874–5.
female adolescent youth in Seattle, Washington, USA. J Adv Nurs 30 Lyerly AD, Little MO, Faden R. The second wave: toward responsible
2002;37:166–72. inclusion of pregnant women in research. Int J Fem Bioethics
196