Child psychiatric disorder and relative age within school year: cross sectional survey of large population sample
BMJ 2003; 327 doi: https://doi.org/10.1136/bmj.327.7413.472 (Published 28 August 2003) Cite this as: BMJ 2003;327:472
All rapid responses
The article by Goodman et al. does not mention the main clinical
issue that clinicians meet regarding children born just before the
watershed of the school year.
When they are born with a gestational age below 32 weeks, they carry
compounded risks of developing psychiatric disorders. One source of risk
is that childhood and adult mental health illness is associated with some
of the same neuro-developmental factors that cause preterm birth. A second
source of risk is that when these biologically immature children enter
school they are allocated to the previous school year not just for a few
weeks or months of difference, but it may be up to 16 months. Finally, the
interaction of environmental (parents) expectations with a child
biologically at risk may become quite complex, depending on a number of
family variables. We are talking of all children born in August and due by
December in the UK and children born preterm at Christmas time in Sweden,
Italy and elsewhere. It is a small percentage in the general population
but it is likely that they are overrepresented in child mental health and
learning disability services.
It is then plausible that if the published study had taken into
account preterm births, stronger correlations could have been found for
this subgroup. On a total population, individual pupils and schools
variability may have shadowed the point that the study could have
contributed to make more clearly.
What the article says, in my view, is that children should be
considered for their individual developmental maturity, rather than be
expected to conform to one-size-fits-all educational policies.
Chronological age is just a proxy measure of biological age. Developmental
issues may not be tackled by administrative shortcuts. Obviously the
latter can help, especially when the role of cognitive aspects (not
directly observable) are overshadowed by behavioural and emotional
manifestations (speaking by themselves).
Competing interests:
None declared
Competing interests: No competing interests
Goodman R, Gledhill J & Ford T have studied almost 10,500 school
children, and they have concluded that younger children in the school year
are more likely to have a psychiatric diagnosis (BMJ 2003;327:472-5).
They say that the youngest third of children in a school year have a
psychiatric diagnosis at the rate of 9.9%, compared to 8.8% for the middle
third and 8.3% for the oldest third in the school year. These percentages
are for their "Any Psychiatric Diagnosis" category.
The statistical significance for these three percentages for "Any
Psychiatric Diagnosis" is P = 0.03 (significant). This is not nearly as
high a level of statistical significance compared to the 0.001 (very
highly significant) which they get for both the parent and for the teacher
reported symptoms for All Ages and for the 5-10 year old group.
Please will Goodman R et al provide us with the breakdown they must
have of this "Any Psychiatric Diagnosis" category they use. Please will
they post on the BMJ website the percentages for each of the three thirds
of a year in each of the three age categories for each and every
psychiatric diagnosis by name.
It is important for the rest of us, especially clinicians, to know
which psychiatric diagnoses were more common and which were less common.
I suggest that the category of "Any Psychiatric Diagnosis" may not be very
helpful, although I suspect that the limitation on space in the journal
may have required this "mixed-bag" category from them.
If the Editor cannot accept a full breakdown of the "Any Psychiatric
Diagnosis" according to the specific DSM4 diagnosis, then please would the
authors email this data to me, and we can then discuss it on bmj.com.
This is an important study and deserves discussion. Please may we
have this data within the next seven days.
Competing interests:
None declared
Competing interests: No competing interests
The study said that: "Teachers often forget to make allowances for a
child's relative age"!.... It appears that they also forget to make
allowances for a child's educational needs, intellectual potential and
already obtained academic ability. Children are not regarded as
individuals in education.
Being the youngest in the year is only associated with educational
disadvantage if the teacher does not make allowances for a child's ability
or needs. Its not the parent that has the unreasonable expectation, its
the system as it expects everybody to function at a certain level at a
certain age in their life regardless of their intelligence, ability, need,
personality or situation.
Grouping children by relative age would only help the solve the
problem if all those children at that particular age were at the exact
same level and stage of their educational, social and emotional
development and have the same needs that need to be met.
Children need to have their particular educational needs met to be
happy and fulfilled in their education. Meeting children’s age
requirements is, in the large majority of cases, not meeting the child’s
educational needs and is very damaging to a child’s self esteem and
feelings of self worth which in turn affects them psychologically and
emotionally.
Surely a more sensible approach would be to start children at school
when they are emotionally and socially ready to start school and then put
the children into classes at a level that is appropriate and suitable to
their ability and need.
Competing interests:
None declared
Competing interests: No competing interests
It was intersesting to read a US perspective on this issue. Here in
Australia, holding back is very much encouraged by education authorities
and the media, and is deemed almost mandatory for boys, who are regarded
as suffering far more disadvantage from "early" school starts.
In my State, the school year begins in late January, and a child may
start if they have turned 5, or will turn 5 by the end of June, but many
are held back until the following year, resulting in a possible age spread
of 18 months within a single class. This situation would appear to put
even more emphasis on the need for teachers to be aware of the differing
ability levels within their classroom. However, it is parents and parental
attitude to schooling which are generaly portrayed as the major cause of
disfunction in children.
The Sydney Morning Herald ran a brief report on this resech paper (It
can be viewed at >
http://www.smh.com.au/text/articles/2003/08/29/1062050664727.htm ) Despite
the fact that authors Goodman, Gledhill and Ford made no mention of
parents as a risk factor in the development of a child's psychiatric
difficulties, the Herald report suggests that "pushy" parents with
unrealistic expectations for their children's academic success are the
primary cause of the disfunction reported in the original paper.
This is quite typical of media attitudes and public perception here.
Thomas Radecki suggested that in the US gifted children who would benefit
from academic acceleration may be held back by school administration. In
Australia, such children are often held back because parents fear the
social disapproval that results from suggesting that a gifted child has
different educational needs.
Unfortunately, it appears that this paper is already on its way to
becoming part of the justification for forcing academicaly gifted children
into lockstep progression with their age peers.
Competing interests:
None declared
Competing interests: No competing interests
Congratulations to Professor Goodman and colleagues (1)on an
excellent demonstration of the way that small effects at an individual
level can create significant opportunities at the population level for
public health interventions. It is possible, as they suggest, that it is
the differential behaviour of teachers towards slightly younger pupils in
the classroom that increases their risk of developing mental health
problems. However, there are other players learning their parts in the
classroom drama: the older children. As professionals and adults it is
easy for us to neglect the longterm impact of early social adversity (2)
but patterns of bullying and sexual aggression can undermine mental
wellbeing in the school age population. At five, if there are going to be
targets for classroom bullying, the younger (smaller, less articulate,
less assertive) child may be at greatest risk. At eleven-plus, the young
people who reach puberty first may show aggression towards their less
developed classmates. Most children do not experience trauma during "the
best years of their lives", but for some The Lord of the Flies can be
closer to their experience of schooldays. Where young people do become
trapped in harmful social situations, it is possible that quite subtle
differences in age and development could increase the risks. However,
this growing knowledge base (1) around child development and life
trajectories could also improve our interventions to promote mental
health.
1 Goodman R, Gledhill J, Ford T. Child psychiatric disorder and
relative age within school year: cross sectional survey of large
population sample. BMJ 2003; 327: 472-475.
2 Caan W. Good for mental health - an academy for the social
sciences. Journal of Mental Health 2000; 9: 117-119.
Competing interests:
Chair of the School Health Research Group, 2001-2003
Competing interests: No competing interests
If the increased level of psychiatric stress found in the relatively
younger students is the result of inappropriate expectations on the part
of teachers or the absence of streaming students by ability, holding back
students may have no net beneficial effect. The held back student will
now become one of the older students in the class making all the other
students relatively younger than they were before and making them appear
of lower ability when compared to the new elevated average. This may
increase the stress on them, and result in no net gain in psychiatric well
-being. Indeed, some held back students may be resentful and increase
stress both in the held back student and in the classroom.
I would point out that at least in the United States, school systems
often have a conflict of interest on the holding back issue, since holding
a child back will very likely increase the number of years that child is
in that particular school system and, therefore, increase the amount of
government funding that school system will receive if there is a capitated
reimbursement system. Since I have personally witnessed students denied
early graduation for just this reason, I would not dismiss it too readily.
Students who are held back are injured in at least one way. They are
stuck in an extremely lengthy educational system one extra year and denied
one extra year of occupational income and freedom from schooling.
A sizeable number of American studies have found no psychological
harm associated with the grade acceleration of the more academically
capable students. Indeed, only one study detected any difference and it
found that gifted students who were not grade accelerated suffered more
psychological difficulties. I fear the BMJ study, with its simply remedy
of holding back, may be used by some less than perfect teachers to hold
back many students unnecessarily. While I applaud the study for better
documenting a problem, I would note that there is absolutely no research
showing that holding back on strictly psychiatric grounds helps any
student, let alone which students. Therefore, I am troubled by its
promotion in the article as a primary remedy. While I think there almost
certainly are psychiatric situations for holding back, we must remember
that harm is also possible.
Unfortunately, here in the U.S., there is often a very politically
correct hostility against streaming students by their abilities and
against grade acceleration. In view of the strong dysgenic effect of
college education on the more academically gifted around the world, we
need to find ways shorten the amount of time college bound students are
forced to spend in school, so that they may have more time to earn money
and have larger families than they currently do. I fear this article will
be used to increase resistance against grade acceleration and to hold
students back simply because they are "more immature," and not because
they have any increased psychiatric difficulties.
Obviously, my 13 year old daughter is more immature physically than
the other freshmen in her college classes, but that doesn't mean she would
have benefited from being held back.
Competing interests:
None declared
Competing interests: No competing interests
Relative age suggests a transactional process
The authors data suggests that the relative age effect is more
apparent for teacher than parent ratings.This would imply a situational
factor in play.
As Prof. Caan points out in his response,the impact of a
constitutional vulnerability such as shyness &/or learning disability
transacting with ordinal position within the class would help define a
"target" population more vulnerable to the negative attention of a
proportion of the older cohort, some of whom will be bullies.
There is no empirical evidence that simply delaying school entry or
repeating the year will address these "relative age" effects.
Clinical experience demonstrates that the solutions for an individual
child will depend on many "bio-psycho-social" factors.
The reduced opportunities for segregated classes for some vulnerable
children have been evident in my practice in recent years.
Whilst the authors present a range of solutions to address the
relative age effect, I think we need to better understand the different
patterns of psychopathology which are not readily visible in the data as
presented. This might suggest different psychopathological mechanisms in
different sub-groups of this large data set.
Laurence Jerome
Competing interests:
None declared
Competing interests: No competing interests