Autism elopement: what the 49% statistic actually means
Elopement is when a child leaves a safe, supervised place without permission. Parents and caregivers usually call it bolting, running away, or wandering off.
You have probably read that 49% of autistic children elope. That number is real, but it does not mean what most pages say it means.
It counted children who tried to leave at least once at any point after their fourth birthday. It is a lifetime count, not a yearly rate, and it came from families who signed themselves up for a research registry.
Where the 49% number comes from
It comes from a 2012 study in Pediatrics that surveyed families of 1,218 autistic children aged 4 to 17. The exact finding was that 49% had attempted to elope at least once after age four.
Two details get dropped almost everywhere that figure is repeated. It counts a child's whole life since their fourth birthday, so it is not a yearly rate. And the families came from an online research registry they joined themselves, which the authors note skews white and highly educated.
None of that makes the number wrong. It makes it a different number than the one most pages are describing.
A better-sampled number almost nobody quotes
In 2016, researchers used the CDC's Pathways Survey, a nationally representative sample of 3,518 children. They asked whether a child had eloped in the previous twelve months.
Across all developmental disabilities, 26.7% had. Among autistic children specifically, it was 32.7%.
| Study | What it asked | Sample | Result |
|---|---|---|---|
| Anderson 2012, Pediatrics | Ever tried to elope after age 4 | 1,218 children, self-selected registry | 49% |
| Kiely 2016, PLoS ONE | Eloped in the past 12 months | 3,518 children, nationally representative | 26.7% overall, 32.7% autism |
These are not competing answers. They are two different questions, and knowing which one you are reading changes how worried the number should make you feel.
What happened to children who went missing
The 2012 study also asked what elopement led to. Of the children who went missing long enough to cause concern, 65% were in danger of traffic injury and 24% were in danger of drowning.
That is a narrower group than the 49%. It describes the children whose elopement turned into a genuine search, not every child who ever tried a door handle.
Water is the risk that comes up most in the research. A 2017 study that screened 32 million US death certificates found drowning among the leading causes of injury death for autistic children.
The numbers you should treat carefully
You will see claims that seven children die each month from wandering, or that 91% of those deaths are drownings. Those come from the National Autism Association's own tracking of news reports about missing children.
That is a record of cases that made the news, not a dataset and not peer-reviewed research. It points in the same direction as the death-certificate study, so the concern behind it is real. It just cannot support a precise rate.
We would rather tell you where a number came from than hand you a scary one with nothing underneath it.
What elopement behavior looks like at different ages
Elopement attempts peaked at age 5.4 years in the 2012 study, with a standard deviation of 2.5 years. That spread matters: the peak is a band roughly from age three to age eight, not a single year.
In a young autistic child it often looks like a straight dash for a door or a gate. In an older child it is quieter, more planned, and easier to miss.
Autistic adults elope too, and it is rarely discussed. The research follows children, so the honest position is that we know far less about autistic individuals after 18 than parents deserve to know.
Elopement prevention: where families start
Most caregivers work in layers. Barriers and door chimes buy time, teaching reduces the reason, and neighbours who recognise the child make an emergency search faster.
Water is the first thing to sort out. After that, families usually deal with the one door their child actually uses rather than the whole house.
Devices such as GPS locators and door alarms come up constantly. Budding Futures does not sell them and will not rank them, but they are worth discussing with your care team and other parents who have tried them.
What Budding Futures does about it
Elopement has a reason behind it, and the reason is different for every child. Our BCBAs run a functional behavior assessment to work out what a child with autism is getting, or getting away from, when they bolt. Sensory overload, anxiety and fear are common drivers, and so is simply wanting something on the other side of the door.
From there the plan teaches a safer way to get the same thing, and builds the supervision and environment around the child while that skill grows. Rachel Blackburn, BCBA, our Clinical Director, oversees that planning.
Nothing here is unique to autism spectrum disorder, or ASD, on its own. Children with ADHD elope as well, and plenty of autistic children never do.
Behavioral treatment for elopement has real supporting evidence, though it is thinner than people assume. The most direct study we can point to is a retrospective review of 11 patients at Emory's Marcus Autism Center, which reported a large effect (Cohen's d = 1.18). Eleven children at one clinic is a small base, and we would rather say that than round it up.
So when a caregiver asks us whether therapy will prevent elopement outright, the honest answer is no. It can reduce how often a child elopes and how dangerous it is, and that is worth a great deal, but it is not the same promise.
Our full approach is on the ABA therapy for elopement page. We work in homes across Colorado and inside schools where the school grants access. ABA reduces how often elopement happens and how risky it is. It does not switch it off, and any provider promising that is overselling.
Frequently asked questions from parents
Q: What is elopement in autism?
A: It is leaving a safe or supervised space without permission. That covers bolting out a front door, running off in a parking lot, and walking away from a classroom.
Q: How common is elopement in autistic children?
A: Common enough that most families of young autistic children deal with it. A nationally representative survey found 32.7% of autistic children had eloped in the previous year. An older registry study found 49% had tried at least once after age four.
Q: At what age is elopement most likely?
A: Attempts peaked at age 5.4 years in the 2012 study, with a standard deviation of 2.5 years. That means the peak covers a wide band of ages, not one birthday.
Q: Is elopement a sign my child is being naughty?
A: No. Elopement almost always serves a purpose for the child, like getting away from something overwhelming or reaching something they want. That is why the first step is working out the reason rather than adding a consequence.
Q: Does elopement stop as a child gets older?
A: It often becomes less frequent, but the honest answer is that the research does not follow individual children far enough to promise that. Safety planning should not assume a child will simply grow out of it.
More on elopement and staying safe
ABA Therapy for Elopement in Colorado
How Budding Futures assesses why a child bolts and builds the safety plan.
Why do autistic children elope?
The four reasons behind bolting, and what an assessment looks for.
How to prevent autism elopement
A layered plan for the home, and an honest word about trackers and alarms.
Sources
Anderson et al., Pediatrics 2012
Occurrence and family impact of elopement in children with autism spectrum disorders. n=1,218.
Kiely et al., PLoS ONE 2016
Prevalence of elopement in a nationally representative sample of children with developmental disabilities. n=3,518.
Guan and Li, Am J Public Health 2017
Individuals with autism at substantially heightened risk for injury death. 32 million death certificates screened.
Call et al., Autism 2017
Clinical outcomes of behavioral treatments for elopement. Emory University, n=11 retrospective case series.
Worried about a child who bolts?
Tell us what happens and when. A BCBA will talk through what the pattern sounds like and what a safety plan would involve.