Journal of Child Psychology and Psychiatry 56:9 (2015), pp 933935
12452
Editorial: Early detection of mental health and
neurodevelopmental disorders: the ethical challenges
of a field in its infancy
The signs of many mental health and neurodevelop-
mental conditions first appear in childhood and
diagnosis can reliably be made by school age for
most. Such conditions can be chronically disabling
and confer significant long-term impairment. Deter-
mining early risk signs and first emerging symptoms
of disorder is imperative to enhance early detection
and to identify targets and ideal time points for
prevention and intervention efforts. It is critical for
diagnostic and classification purposes to identify
early behavioral patterns that are disorder-specific.
In addition, identifying trans-diagnostic markers
that overlap across disorders could enhance the
impact of prevention and early intervention efforts
(Nolen-Hoeksema & Watkins, 2011; see also a recent
JCPP Editorial1 on identifying core dimensions of
early functioning).
This Special Issue of JCPP focuses on the prospect
of earlier identification of conditions that are tradi-
tionally diagnosed later in childhood. Ten invited
contributions cover topics related to the science of
early detection. Several are focused on prediction:
of later diagnosis, of functional impairment, and of
future service utilization. Sullivan and colleagues2
explore the earliest developmental period, examining
whether identifiable patterns exist at 6 months of
age that predict ADHD risk. Rajendran et al. and
Thompson et al. prospectively track high-risk sam-
ples annually from age 3 to later childhood, exam-
ining predictors of ADHD and dyslexia, respectively.
Dougherty and colleagues study preschool irritability
and its trans-diagnostic associations with a broader
range of psychopathology, as well as functional
impairment, at age 9. Chorozoglou and colleagues
examine the predictive ability of another symptom
detectable in preschool, hyperactivity, and its asso-
ciation with later economic burden and service
utilization. Other papers focus on instrument devel-
opment (Mian et al., Petitclerc et al.), early screening
for the purposes of widespread preventive interven-
tion (Mihalopoulos and colleagues), and screening
and diagnostic accuracy and stability (Ozonoff et al.;
Sheldrick et al.).
The articles in this Special Issue demonstrate
multiple benefits of early detection. Identification of
delays and symptoms prior to school entry may
prevent academic failure, school behavior problems,
dropout/delinquency, and the development of more
severe mental health issues in later childhood,
adolescence, and adulthood. Early detection has
the potential to decrease service utilization and
economic burden later in life. The plasticity of the
2015 Association for Child and Adolescent Mental Health.
Published by John Wiley & Sons Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main St, Malden, MA 02148, USA
doi:10.1111/jcpp.
brain in the first three years of life suggests that
intervention at this time will be most effective,
further highlighting the urgency of early identifica-
tion. It is important, however, to take a balanced
look at early detection,
What we choose to
weighing such benefits
publish, how we inter-
against potential ethical
risks as well.
pret our data, and the
One of the clearest clinical implications we
ethical challenges is draw are, themselves,
the uncertainty of test ethical choices
results and the risk of
false-positive diagnoses at a very young age. The
early years of life are characterized by rapid
changes in development as well as significant
behavioral variability from moment to moment,
making screening and diagnosis challenging. Psy-
chometric tools with adequate measurement prop-
erties (high sensitivity, specificity, and positive
predictive value) do not exist for many conditions
first evident during the toddler and preschool
years. Inaccurately identifying a child as having a
neurodevelopmental or mental health diagnosis
can cause a host of negative downstream effects.
False-positive early identification arouses unneces-
sary anxiety in parents. In the field of mandatory
newborn screening for genetic disorders, these
potential risks have been well articulated, includ-
ing negative effects on the parent-child relation-
ship, attachment, parental attitudes, and family
dynamics (Hewlett & Waisbren, 2006). Anxiety may
remain, even after an early diagnosis is shown to
be a false positive, with parents continuing to see
the child as at risk.
Incorrect early identification may also lead to the
recommendation of unnecessary treatments that
may be costly and potentially invasive (e.g.,
unneeded medications). Depending upon the state
of the science, the child may be exposed to interven-
tions that are as yet untested. Research on very early
detection of many disorders is still maturing and
there is a consequent paucity of validated treatments
for infants and toddlers. Our field is caught in a
tautology that is likely to continue for some time and
may be unavoidable when a research area is matur-
ing: until we are accurately identifying infants,
toddlers, and preschoolers with developmental and
psychiatric challenges, we cannot conduct research
to develop and validate age-appropriate interven-
tions. Clinicians must be keenly aware of the limits of
current science and the ethical issues they raise and
communicate these to families in a clear manner.
934 Editorial: early detection
False negatives also pose a major ethical chal-
lenge. It is no better, and perhaps worse, to inaccu-
rately tell a parent that their childs development is
fine, only to later find out that a diagnosis, and the
opportunity for early intervention, were missed.
The first step in minimizing these ethical chal-
lenges is to apply rigorous science to the identifica-
tion of early risk markers, so that we have a solid
understanding of their base rates in the population
and their sensitivity, specificity, and positive and
negative predictive values in high risk groups, as
articulated by Sheldrick and colleagues. Several
papers in this Special Issue are devoted to identifi-
cation of earliest risk markers (Dougherty et al.;
Sullivan et al.; Thompson et al.), evaluation of psy-
chometric indices of early prediction (Ozonoff et al.;
Sheldrick et al.), and development of better measure-
ment tools for very young children (Mian et al.;
Petitclerc et al.). These papers advance our under-
standing of the earliest red flags for different disor-
ders, how to measure them validly, and the kinds of
mistakes that may be made in applying predictive
algorithms at very young ages.
For some disorders, research on early markers has
matured to the stage that there is clear empirical
evidence of a prodrome that precedes the onset of
frank symptoms. In schizophrenia, for example, a
great deal has already been written about the ethical
issues of identifying individuals who are asymp-
tomatic or subthreshold in expression of later-onset
disease (Corcoran, Malaspina, & Hercher, 2005). We
need to better understand both the benefits and the
costs of identifying such patients-in-waiting
(Golden-Grant, Merritt, & Scott, 2015), especially
when preventive interventions have not yet been
developed. All the caveats just raised about the risks
of false positives and negatives and the limited
toolbox of appropriate measures apply to identifica-
tion of pre-symptomatic markers as well. For most
diagnoses, there are not clear boundaries between
normality and disorder, and this will likely be even
truer of prodromal markers, leading to the risk of
pathologizing a minor variant or behavior that is
consistent with typical development. Furthermore, it
is not clear whether the parents of infants or toddlers
who screen positive for a disorder, or for a marker of
impending disorder, will choose to access further
clinical services. For example, the rate of compliance
with advice to pursue additional evaluation, after
screening positive for possible ASD at 14 months, is
relatively low (Dietz, Swinkels, van Daalen, van
Engeland, & Buitelaar, 2007). How invested will
parents be in pursuing further medical attention
when a young child is not yet (very) symptomatic?
So where does this leave us? From a research
perspective, it is clear that studies of early detection
must forge ahead swiftly, in order to improve and
mature the state of the science and permit the
development and testing of early treatments and
preventive interventions.
J Child Psychol Psychiatr 2015; 56(9): 9335
But, from a clinical perspective, what are we to do
to help our patients and their families? First, we
must clearly convey to parents the limits of our tests
and our ability to predict outcomes. Families will
vary in their tolerance for ambiguity, but it is
nonetheless important to express that we are
currently better at predicting risk than making a
definitive diagnosis at these very early ages. There is
a burgeoning literature on risk communication and
a working group has put together recommendations,
applicable across childhood disorders, on how to
communicate the difference between risk and diag-
nosis to parents (Yudell, Tabor, Dawson, Rossi,
Newschaffer, & Working Group in Autism Risk
Communication and Ethics, 2013). Risk markers
increase the likelihood of a diagnosis, but are not the
same as a diagnosis. The concept of risk includes
the uncertainty that exists in the current state of
early detection science and measurement. Being
open and transparent with parents about this
uncertainty will promote trust and encourage long-
term engagement with the clinical care system, even
if the initial interpretation of the childs status
changes. We, and our science, can help parents
understand the range of outcomes, provide hope for
the future, and address developmental concerns
with practical advice applicable to children at risk,
even when early uncertainty exists about a specific
diagnosis (Caronna, Augustyn, & Zuckerman,
2007).
Finally, it is imperative that scientists are always
aware of the limitations of our studies. What we
choose to publish, how we interpret our data, and
the clinical implications we draw are, themselves,
ethical choices (Rysavy & Murph, 2015). Our work
has great power to influence families. We must wield
this power carefully, always striving to help parents
make informed decisions about the best interests of
their child, as we move science forward.
Sally Ozonoff
JCPP Joint Editor and Special Issue Editor
Acknowledgement
S.O. (a joint Editor of JCPP) acted as Guest Editor of
this Special Issue; she has declared that she has no
competing or potential conflicts of interest in relation to
the Editorial or Special Issue, for which all the articles
were subject to regular peer review and revision; the
peer-review decision-making for the article by Ozonoff
et al. was handled by the JCPP Editor-in-Chief (Ed-
mund Sonuga-Barke). S.O. has no conflicts declared in
her capacity as JCPP Joint Editor.
Notes
1. JCPP Editorial: Sonuga-Barke, E. J. S. (2014),
Editorial: Whats up, (R)DoC? can identifying
core dimensions of early functioning help us
understand, and then reduce, developmental
2015 Association for Child and Adolescent Mental Health.
doi:10.1111/jcpp.12452
risk for mental disorders? Journal of Child
Psychology and Psychiatry, 55, 849851.
2. Articles that are in this issue (with itali-
cised names on first mention) can be found at:
http://onlinelibrary.wiley.com/doi/10.1111/jcpp.
2015.56.issue-9/issuetoc.
References
Caronna, E.B., Augustyn, M., & Zuckerman, B. (2007).
Revisiting parental concerns in the age of autism spectrum
disorders: The need to help parents in the face of uncertainty.
Archives of Pediatric and Adolescent Medicine, 161, 406408.
Corcoran, C., Malaspina, D., & Hercher, L. (2005). Prodromal
interventions for schizophrenia vulnerability: The risks of
being at risk. Schizophrenia Research, 73, 173184.
Dietz, C., Swinkels, S.H.N., van Daalen, E., van Engeland, H.,
& Buitelaar, J.K. (2007). Parental compliance after
screening social development in toddlers. Archives of
Pediatric & Adolescent Medicine, 161, 363368.
2015 Association for Child and Adolescent Mental Health.
Editorial: early detection 935
Golden-Grant, K., Merritt, J.L., & Scott, C.R. (2015). Ethical
considerations of population screening for late-onset genetic
disease. Clinical Genetics. Advanced online publication. doi:
10.1111/cge.12566.
Hewlett, J., & Waisbren, S.E. (2006). A review of the
psychosocial effects of false positive results on parents
and current communication practices in newborn
screening. Journal of Inherited Metabolic Disease, 29,
677682.
Nolen-Hoeksema, S., & Watkins, E.R. (2011). A heuristic for
developing transdiagnostic models of psychopathology:
Explaining multifinality and divergent trajectories.
Perspectives on Psychological Science, 6, 589609.
Rysavy, M.A., & Murph, J.R. (2015). On risks and reality:
Communicating the difference between autism risks and
diagnosis. AMA Journal of Ethics, 17, 323327.
Yudell, M., Tabor, H.K., Dawson, G., Rossi, J., Newschaffer, C.,
& Working Group in Autism Risk Communication and
Ethics. (2013). Priorities in autism spectrum disorder risk
communication and ethics. Autism: The International
Journal of Research and Practice, 17, 701722.