Age of diagnosis for autism: individual and community factors across 10 birth cohorts

6 Nov

Prof. Peter Bearman’s group at Columbia has a new paper out: Age of diagnosis for autism: individual and community factors across 10 birth cohorts.

The study looks at children receiving services from the California Department of Developmental Services (CDDS). The CDDS dataset is not a “census” of autistics in California, but is a registry of those who have sought services, have been identified as autistic and and who have been granted services. Factors which can affect who seeks services for their children, who seeks and how easy it is to identify an autistic child (e.g. access to people who can do the diagosis) will have an effect on who is identified and when that person is identified.

As an aside, I focus on autistic children here because the study does. The authors focused on those identified who were under age 8. They did this to make the comparisons consistent across birth cohorts. For example, one can’t look at 15 year olds born in the year 2000.

Age of diagnosis was taken as age of autism into the entry for the child into the CDDS registry. This can occur either when the child undergoes an intake or a change in the child’s status occurs (say, a diagnosis of autism is given to a child already in the CDDS system)

California statutorily requires that regional centres confirm eligibility for services, including verifying or conferring a diagnosis, within 120 days of intake, so the date at which DDS clinicians either provided or confirmed a first diagnosis of autism is within a few months of caregivers’ initial request for assistance. Combined with the child’s date of birth, we then used the date of entry from the first available CDER to calculate the age of diagnosis. Diagnoses earlier than age 3 years are empirically rare, as infants and toddlers below 36 months with suspected developmental delays and those considered at risk are served by the early start programme.

I don’t have the data, but my very anecdotal and likely biased experience is that currently diagnoses earlier than 36 months are not rare at all.

Prof. Bearman’s group shows that the age of diagnosis steadily dropped during the 1990’s, from 4.4 years of age in 1992 to below 3.4 years of age in 2002. Here’s the figure from the paper (click to enlarge):

Non-White and Hispanic children were diagnosed later. Children of highly educated parents were diagnosed earlier. Both of these effects remained throughout the time span considered (1992-2002). Children of mothers born outside the US and first born children were diagnosed later, but this effect disappeared over time. Children with better communication skills were diagnosed later.

They also found that the age of diagnosis depends on the parents socio-economic status (ses). As one might predict, better off families got diagnoses for their children earlier. Here is the figure showing the trend of age-of-diagnosis vs. birth year for low and high ses:

It is good to see the gap decreasing with time, but it shouldn’t be there at all.

One obvious question that comes up from this study–a question that it can not answer–is how many people are never correctly diagnosed, and what gap might there be in that number based on ses?

The full paper can be found on Prof. Bearman’s team website. Here is the abstract:

Background The incidence of autism rose dramatically between 1992 and 2001, while the age at which children were first diagnosed declined. During this period the size and composition of the autism caseload has changed, but little is known about whether the factors associated with the timing of diagnosis may also have shifted. Using a multilevel analysis strategy, the individual and community-level factors associated with age of diagnosis were modelled across 10 birth cohorts of California children.

Methods Linked birth and administrative records on 17?185 children with diagnoses of autistic disorder born in California between 1992 and 2001 and enrolled with the California Department of Developmental Services (DDS) were analysed. Information on cases, their parents and their residential location were extracted from birth and DDS records. Zip codes of residence were matched to census data to create community-level measures. Multilevel linear models were estimated for each birth cohort, with individual-level effects for sex, race, parental characteristics, poverty status, birth order and symptom expression. At the community level measures of educational and economic composition, local autism prevalence and the presence of a child psychiatrist were included.

Results Children with highly educated parents are diagnosed earlier, and this effect has strengthened over time. There is a persistent gap in the age of diagnosis between high and low socioeconomic status (SES) children that has shrunk but not disappeared over time.

Conclusion Routine screening for autism in early childhood for all children, particularly those of low SES, is necessary to eliminate disparities in early intervention.

Here is the press release for the study.

Autism is diagnosed later for children with less educated parents
A Columbia study, appearing this week in the Journal of Epidemiology and Community Health, has found a gap in age of diagnosis for autism between children of high and low socioeconomic status in California. This gap has become smaller over time, falling from about fourteen months to about six months in a decade, but it remains significant. The strongest factor in this gap was parental education. Children of highly educated parents tend to be diagnosed at earlier ages, and this effect has not diminished over time.

The findings suggest that although the median age of diagnosis for autism has dropped from about four and a half years in the early 1990s to about three and a half in the 2000 birth cohort, there are some groups of children who are still diagnosed late. Diagnosis is the crucial first step to treatment, widely believed to be most effective at younger ages, and even six months may be important at an age when children are developing rapidly. “Our findings point strongly to the idea that some children may be at a great disadvantage when it comes to access to diagnosis and treatment for autism,” said Peter Bearman, the Jonathan Cole Professor of the Social Sciences and principal investigator of the paper. “These delays may have important consequences for later behavioral and cognitive outcomes.”

Autism impairs social interaction and predisposes children to restrictive and repetitive behaviors. Over the past two decades California has witnessed a particularly large spike in autism cases. Between 1992 and 2006, the state’s caseload increased 598 percent. At the same time, the typical age of diagnosis has dropped from school-age to the early pre-school years.

This study was based on 17,185 children with autism born in California from 1992-2001. In addition to looking at characteristics of the children and their parents, from the birth records, the researchers used the zip code at birth and diagnosis to examine the characteristics of the communities in which the children lived.

The researchers found that children born to less-educated mothers, and those whose births were paid for by Medi-Cal (California’s Medicare program) were diagnosed later. In addition, non-White and Hispanic children, and those with mothers born outside the US, showed delayed diagnosis. In the early part of the decade, firstborn children were also diagnosed later, suggesting that parental familiarity with typical child development may have been a factor, however this effect disappeared over time as autism awareness spread.

The kinds of neighborhoods where children lived also mattered, particularly in the early years of the study. In these years, children living in areas that had many children with autism were diagnosed early, which indicates that familiarity with the symptoms of autism may have been important. Children born in neighborhoods with higher property values were also diagnosed earlier. In general though, over time the importance of neighborhood characteristics seems to have diminished, perhaps because autism became more visible and recognized.

“The findings suggest that for many children, increasing awareness of autism and regular screening has succeeded in indentifying cases of autism at earlier ages,” said Christine Fountain, postdoctoral researcher and lead author of the paper. “However we need a better understanding of how information about autism spreads between parents, teachers, and physicians, and how parents marshal their resources to obtain diagnoses and services for their children in a timely way. This will help us to make sure that some children aren’t left undiagnosed and without the help that they need.”

The study was supported by a National Institutes of Health Pioneer Grant, given to scientists pursuing new strategies to improve health, and was conducted through Columbia’s Paul F. Lazarsfeld Center for Social Sciences. More information on this study can be found at

5 Responses to “Age of diagnosis for autism: individual and community factors across 10 birth cohorts”

  1. David Cole November 6, 2010 at 02:55 #

    The report states:

    “Diagnoses earlier than age 3 years are empirically rare, as infants and toddlers below 36 months with suspected developmental delays and those considered at risk are served by the early start programme.”

    You respond:

    “I don’t have the data, but my very anecdotal and likely biased experience is that currently diagnoses earlier than 36 months are not rare at all.”

    To clarify:

    In California, children younger than 3 years of age who display some symptoms of developmental delays or are deemed at risk, due to a family history of developmental delays, are eligible for services in the absence of a qualifying diagnosis (via the mentioned Early Start Program).

    While some children younger than 3 years of age still receive diagnoses from private providers, (1) the motivation for caregivers to pursue a diagnosis privately is reduced, at least until the children near the age of 3, and (2) the CDDS is not inclined to diagnose children who already qualify for its services when the ostensible benefit of lowering the criteria for services prior to 3 years of age is that many children who would have been diagnosed with a developmental delay (namely, Autism) in the absence of intervention no longer meet the diagnostic criteria when the CDDS or a private provider assesses them at a later age.

  2. livsparents November 6, 2010 at 17:31 #

    Legislation, the other Occam’s Razor:

    The California State Legislature passed the California Early Intervention Services Act in 1993 (CEISA; 14 G.C. Section 95000 et seq.). CEISA established state authority to develop an early intervention service system that was congruent with federal requirements. The California Department of Developmental Services (DDS) plans, develops, implements, and monitors the statewide early intervention services system in collaboration with the California Department of Education (CDE),and with advice and assistance from the State Interagency Coordinating Council (ICC). The Departments of Health, Social Services, Mental Health, and Alcohol and Drug Programs cooperate and coordinate with DDS and CDE in the delivery of early intervention services.
    State regulations governing Early Start were approved in August 1998 and are found in Title 17, California Code of Regulations, Section 52000 through 52175.

    if you legislate it…they will come…

  3. Barbara November 8, 2010 at 12:31 #

    Sullivan, it’s not the main point you make, and this may be tangential but you write about anecdotal evidence suggesting that the diagnosis of autism pre age 36 months is not particularly rare. There is also some research evidence for this, dating from the 1970s (I think historical perspectives are sometimes useful as to what they reveal)

    Lovaas’s entire hypothesis was based on his experimental cohort being diagnosed, and ‘treated’ to his intensive interventions, before the age of 36 months. In fact, one of the many flaws in that dreadfully flawed paper was that his main control group was 6 months older on average than his experimental group. He used also as a comparator a group of autistic children working with another researcher, Freeman and his team. Freeman’s team was noting changes in various elements of adaptation, and found that the symptoms and behaviours and the ‘IQ’ of children diagnosed with autism were fairly stable over time – except in one group. That group was children aged 3 or under at the time of diagnosis. Several of these made sudden gains, a couple regressed. A diagnosis at/before age 3 appeared, even in the 1970s, to this group of researchers, to be highly unreliable. Even 6 months later made a lot of difference.

    So California, or at least that little UCLA enclave, was diagnosing autism pre 36 months in the 1970s, enough cases for several teams of researchers to work on. And the diagnosis was maybe unreliable. Lovaas must have known that, mustn’t he?


  1. Tweets that mention Autism Blog - Age of diagnosis for autism: individual and community factors across 10 birth cohorts « Left Brain/Right Brain -- - November 6, 2010

    […] This post was mentioned on Twitter by Brandon Blietz, Alltop Autism. Alltop Autism said: Age of diagnosis for autism: individual and community factors across 10 birth cohorts […]

  2. Autism Blog – Age of diagnosis for autism: individual and … | My Autism Site | All About Autism - November 6, 2010

    […] See original here: Autism Blog – Age of diagnosis for autism: individual and … […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: