change the world badge

change the world badge


Child Psychology Blogs

Concerned About Unconventional Mental Health Interventions?

Concerned About Unconventional Mental Health Interventions?
Alternative Psychotherapies: Evaluating Unconventional Mental Health Treatments

Friday, October 29, 2010

When is a Five-Year-Old Older Than Another Five-Year-Old?: Interesting News About ADHD

It wouldn’t surprise most of us if a baby of 6 months acted a lot different from a one-year-old, and a one-year-old a lot different from a two-year-old. Those age differences eventually smooth out, of course, so someone who is 23 probably doesn’t behave much differently from someone who is 22. But when does that smoothing-out occur? We usually treat school-age children as if their grade level is more significant than their actual age in years and months, but this may be a mistake. A recent health economics research article suggests that it may be wrong to assume too early that exact age is irrelevant ( Elder, T. [2010]. The importance of relative standards in ADHD diagnosis: Evidence based on exact birth dates. Journal of Health Economics, 29, 641-656.

Elder’s work used a group of almost 12,000 children and examined the proportion of children diagnosed with attention deficit/hyperactivity disorder (ADHD) in groups that were the oldest in their grades or the youngest in their grades. Because public schools have an age/ birthday cut-off for school entrance and enforce this strictly, looking at a child’s birth date tells us whether he or she entered public school soon after the required birthday and thus was among the youngest in a class, or did not enter until almost a year after that birthday and was among the oldest in the class. If the cut-off date was September 1, children with August birthdays would enter kindergarten at no more than 5 years and one month, but a September birthday would mean that the child waited a year for school entry and came in at no less than 5 years and 11 months of age. Both August and September-born children would be 5 years of age, but the oldest children would be nearly 20% older (and more developed) than the youngest children.

Elder’s study showed that in fact the nearly-20% age difference had significant effects on the children’s ability to cope with the demands of school. About 10% of the “youngest” children (with birthdays shortly before school entrance) were diagnosed as having ADHD, compared to only 4.5% of the “oldest” children (birthdays almost a year before school entrance). In addition, there are long-term consequences of this diagnosis, and by fifth grade almost twice as many of the “youngest” children were taking stimulant medications intended to treat their attention problems.

This pattern was the same for states with a late-summer cut-off as they were for states that set their cut-off later. In all cases, younger children within a group were more likely to be diagnosed and treated for attention problems than older children were. Looking at teachers’ and parents’ assessments of children, Elder found that teachers were very likely to evaluate younger children as having more attention problems, as they compared them to other children in their class; parents were less likely to do this.

Elder’s work suggests that evaluation of attention problems in kindergarteners is influenced by the child’s maturity relative to that of other children in a group. An accurate diagnosis for a child of this age requires comparison to others who are close in age, not to those who are 10, 15, even almost 20% older than the child in question. A child who is 5 years and one month old is as much less mature than one who is 5 years and 11 months old, as a 13-year-old is in comparison to a 15-year-old. Teachers need to be aware of these facts, as their recommendations and evaluations can play a major role in diagnosis of children.

Elder’s work is of great importance in understanding diagnosis of ADHD, one of the most common mental health diagnoses for children ( It is also relevant to the frequent use of medications that have the potential for adverse side effects, and to the creation of adverse expectations in adults who are aware of a child’s diagnosis-- both situations that may worsen the child’s developmental outcome.

The demonstration of relative-age factors for ADHD should also serve as a red flag with respect to various on-line checklists offered for diagnosis of other problems. For example, at, a checklist is presented, purporting to be appropriate for children from age 5 through teen-age. It includes the item “has frequent or intense angry outbursts” without noting that what is “frequent” or “intense” for a 5-year-old may be quite different from what is defined in those terms for a teen-ager.

When making decisions about what a child needs, the question “how old is the child?” is always of primary importance. School systems show that they understand something about this by setting age cut-offs for school entry, but teachers, parents, and professionals need to remember that the details of age can make a big difference in a child’s needs and abilities. One 5-year-old can be a lot older than another 5-year-old.


  1. Makes sense to me and is a very good thing to study. Details of age could also include factors such as birth order, number of siblings, how early they read, hmmmm...I bet size of the child could factor in as well, couldn't it?

    Quite often see higher expectations of extra tall children although one would hope knowing the actual age of kids with stand out height would curb the tendency to think of them as older than they are but I don't think it does always.

  2. Aha! My youngest son, born in July with an Oct cut-off for school, was referred to the child study team and repeated kindergarten, something he always resented. There was no firm diagnosis of ADHD and no medication, but the whole thing was treated as something wrong with the child, not the system.

    He is now a very successful accountant so I guess he outgrew the problem:-)

  3. I have seen literature to the effect that this is apparent longterm and over a range of outcomes. For example, one good reason for getting rid of the 11 plus exam in the UK was that children who passed were on average several months older. The effect even works with premier league footballers who are more likely to be born at the early end of the academic year. It may be that this is particularly pronounced in the UK wehere children start school the year they are 5. This means a child can start scholl aged 4 years and a day, in the same class as children who area ll but 5.

  4. There's a lot about improved outcomes for relatively older children, but not so much until recently about specific diagnostic decisions about the younger ones. But in either case you can see how a cascade of effects can follow the initial step.

    What about gender differences? Did more girls than boys pass the 11 plus, as more boys than girls get the ADHD label?