The Adverse Childhood Experiences Study is the piece of research that put childhood adversity on the public-health map. It was not designed to do that. It began as an attempt to solve a puzzle in a weight-loss clinic, and it ended up changing how a great many doctors, teachers and social workers think about the histories of the people in front of them.
An accident in an obesity clinic
In the mid-1980s, physicians running an obesity programme at a large California health system noticed something strange. The programme worked — patients lost substantial weight — but a striking number of the most successful participants dropped out. People who had lost a hundred pounds or more were the ones most likely to leave.
Following that thread through patient interviews turned up a pattern nobody had gone looking for: a high proportion of those patients had experienced sexual abuse, violence or profound instability as children, and for some of them weight had become protective rather than accidental. That observation was too big for one clinic, and it prompted a far larger question. If childhood adversity was shaping adult health this powerfully, how common was it in the general population?
The study itself
Between 1995 and 1997, the health system partnered with the Centers for Disease Control and Prevention to find out. More than 17,000 adults — mostly middle-class, mostly insured, mostly middle-aged, and overwhelmingly people who had come in for a routine physical exam — completed a confidential questionnaire about their childhoods alongside a standard medical evaluation.
That population matters. This was not a study of a marginalised group, a clinical sample or a prison population. It was a study of ordinary insured adults in comfortable circumstances, which is precisely why the results were so difficult to dismiss.
The questionnaire asked about ten specific categories of experience before the age of eighteen, grouped into abuse, neglect and household dysfunction. We describe all ten in ordinary language on the ten categories. Each category a respondent had experienced counted as one point, producing a score from zero to ten.
What the results showed
Two findings came out of the first analysis, published in 1998, and both have held up across hundreds of subsequent studies in other countries and other populations.
The first is that adverse childhood experiences are common. Roughly two-thirds of this comfortable, insured sample reported at least one. They also cluster: someone who reports one is considerably more likely than average to report several, because the conditions that produce one adversity tend to produce others.
The second is that the relationship between adversity and adult health is graded. As the score rises, so does the statistical likelihood of a long list of adult outcomes — depression, substance use, heart and lung disease, liver disease, and early death among them. It is a dose-response curve of the kind epidemiologists usually see with tobacco or lead exposure, and it appears across outcomes that seem to have nothing to do with each other.
Why one score predicts so many different things
The mechanism has two strands, and both are needed to explain the data.
The first is biological. Prolonged, severe stress in early childhood, without the buffering of a reliable adult, alters the systems that regulate stress hormones, immune response and the developing architecture of the brain. That is the phenomenon researchers call toxic stress, and it is why the effects show up in organs that have no obvious connection to psychology.
The second is behavioural. Some of the coping strategies that make an unbearable childhood survivable — smoking, drinking, eating, dissociating — carry their own long-term costs. The study's authors made a point of this: these behaviours often begin as solutions, and treating them purely as problems misses the reason they exist.
What the study cannot tell you
The ACE Study is strong evidence about populations and weak evidence about individuals, and conflating the two causes real harm.
- It is not a diagnostic instrument. The ten questions were written to characterise a population, not to assess a person. No score establishes any condition.
- It is not predictive for an individual. Elevated risk across ten thousand people says nothing definite about any one of them. Plenty of people with high scores are healthy; plenty with scores of zero are not.
- It was not exhaustive. The ten categories omit a great deal that harms children — poverty, racism, bullying, community violence, medical trauma, the death of a parent, foster placement, natural disaster. Their absence reflects the questionnaire's origins, not their importance.
- It measured presence, not severity or duration. A single incident and years of repeated harm count identically. So does an experience that was met with immediate support and one that was met with none.
Our page on what a score does and does not mean works through those limits in more detail, because they are where the popular version of this research most often goes wrong.
Where the research went next
The CDC folded ACE questions into its ongoing state health surveys, which is how we now have national estimates rather than a single clinic sample. A later CDC analysis estimated that preventing adverse childhood experiences could substantially reduce the national burden of depression, heart disease and several other conditions.
Attention has also shifted from measurement toward prevention and buffering — what actually protects children, and what communities can realistically do. The CDC continues to maintain a public resource on both. That is the more useful half of the story, and it is the subject of most of the rest of this site.