Adverse Childhood Experiences (ACEs) and Long-Term Development

The original ACE Study, conducted by the CDC and Kaiser Permanente between 1995 and 1997, changed the way medicine and public health think about childhood stress — not as a temporary emotional state but as a biological event with measurable consequences that extend decades into adult life. This page covers the definition and scoring of ACEs, the neurobiological and social mechanisms through which early adversity shapes development, how ACEs are classified and contested in research, and what the evidence actually says about resilience and recovery. The science is more nuanced than the headlines — and more hopeful, too.

• Definition and scope
• Core mechanics or structure
• Causal relationships or drivers
• Classification boundaries
• Tradeoffs and tensions
• Common misconceptions
• ACE Score: What the 10 Questions Cover
• Reference table: ACE categories and associated long-term outcomes

Definition and scope

An Adverse Childhood Experience is a potentially traumatic event occurring before age 18 that falls into one of three broad domains: abuse (physical, emotional, or sexual), household dysfunction (substance abuse, mental illness, incarcerated household member, domestic violence, or parental separation/divorce), and neglect (physical or emotional). The original CDC-Kaiser study identified 10 discrete categories — not individual events but categories — meaning a child who witnesses domestic violence twice still scores 1 on that dimension, not 2.

The study's founding sample was not what anyone would call a hard-luck population. Of the approximately 17,000 adults surveyed, more than 70% were college-educated and white — making the results, if anything, a conservative estimate of ACE prevalence across the general US population (CDC, About the CDC-Kaiser ACE Study).

ACE scores run from 0 to 10. Roughly 64% of adults in the original study reported at least one ACE; 12.6% reported 4 or more (CDC-Kaiser ACE Study findings). That second number — the 4+ cluster — became a landmark in public health, because the dose-response relationship between ACE score and health outcomes is one of the most replicated findings in the literature.

Core mechanics or structure

The architecture of ACE effects runs through three systems: the hypothalamic-pituitary-adrenal (HPA) axis, the immune system, and the developing brain's structural wiring. When a child experiences repeated stress without adequate adult buffering, the HPA axis produces cortisol in patterns that, over time, alter gene expression, disrupt myelination of prefrontal circuits, and reduce hippocampal volume — the part of the brain most associated with memory and stress regulation (National Scientific Council on the Developing Child, Harvard University).

Harvard's Center on the Developing Child distinguishes three types of stress response: positive (brief, mild), tolerable (more serious but buffered by a caring adult), and toxic (prolonged, severe, and unsupported). It is the toxic stress pathway — not adversity per se but unsupported adversity — that drives long-term neurobiological change. This distinction matters enormously for trauma-informed approaches to child development, where the presence or absence of a protective adult relationship is treated as a direct biological variable, not just an emotional comfort.

The immune dysregulation piece is less widely known but equally documented. Elevated inflammatory markers — including C-reactive protein and interleukin-6 — have been found in adults with high ACE scores, suggesting that childhood stress literally reshapes immune signaling in ways that contribute to cardiovascular disease, autoimmune conditions, and metabolic disorders decades later (ACES Too High News, citing Felitti et al. original study in American Journal of Preventive Medicine, 1998).

Causal relationships or drivers

The epidemiological signal from the original study was striking enough that researchers spent years trying to rule out confounders. The dose-response relationship held after controlling for socioeconomic status, race, and education. Adults with ACE scores of 4 or more showed:

• 2× the risk of heart disease compared to those with an ACE score of 0
• 7.4× the risk of alcoholism
• 12.2× the likelihood of attempted suicide

(Felitti et al., American Journal of Preventive Medicine, Vol. 14, Issue 4, 1998)

These aren't correlations with trivial effect sizes. The biological pathway is coherent: toxic stress dysregulates cortisol, which impairs executive function development in children, which in turn increases risk-taking behavior in adolescence, which feeds into substance use, which then compounds health risk through direct physiological damage. The risks compound through multiple layers rather than operating through a single mechanism.

Poverty intensifies these effects but does not explain them. A household experiencing domestic violence at a high income level still produces neurobiological stress in children. The resource question matters for recovery and buffering, but the mechanism itself is psychological safety, not material security. Understanding the relationship between economic hardship and ACEs is covered in more depth at poverty and child development.

Classification boundaries

The original 10-category ACE framework has been both celebrated and criticized for what it includes and excludes. What it excludes is significant: community violence, bullying, racism, foster care placement, natural disasters, and death of a parent or sibling are not part of the original scoring. The Philadelphia ACE Survey, developed by researchers at the Robert Wood Johnson Foundation, expanded the framework to include community-level adversity — neighborhood violence, discrimination, unsafe housing — and found that these "expanded ACEs" independently predicted poor health outcomes even when controlling for household ACEs.

The World Health Organization uses a broader adverse childhood experiences framework in its global surveys, recognizing that the Kaiser-CDC instrument reflects a particular American, predominantly white, middle-class baseline (WHO, Violence Against Children).

Within the original 10, the categories are treated as equivalent — a score of 4 counts the same whether it's composed of neglect and household mental illness or physical abuse and incarcerated parent. Most researchers acknowledge this is a methodological convenience, not a claim that all ACE types carry identical neurobiological weight.

Tradeoffs and tensions

The ACE framework sits in a productive but uncomfortable tension between biological determinism and the politics of resilience. The dose-response data is real; so is the inconvenient fact that most people with high ACE scores do not develop chronic disease, addiction, or mental illness. In the original study, roughly 30% of adults with ACE scores of 4 or more reported no adverse health outcomes as measured by the study's instruments.

This creates a dual risk in communication. Overemphasizing the deterministic framing can pathologize children who've experienced adversity and reduce clinician and parent expectations in ways that themselves become a developmental hazard. Overemphasizing resilience can minimize real biological risk and provide cover for inadequate policy responses. Attachment theory and child development literature threads this needle most carefully, locating resilience not as an individual trait but as a relational one — the product of at least one stable, responsive relationship rather than an intrinsic property the child either has or doesn't.

There is also a tension in the screening debate. The American Academy of Pediatrics has acknowledged ACE screening in pediatric settings as a concept worth studying, while stopping short of universal mandates because the research on clinical outcomes from screening — as opposed to research on ACE prevalence and effects — remains incomplete as of the AAP's 2021 policy positions (American Academy of Pediatrics, Pediatrics journal).

Common misconceptions

Misconception: A high ACE score predicts a bad outcome. It predicts elevated population-level risk. At the individual level, the relationship is probabilistic, not deterministic. Neuroscience researcher Bruce Perry and psychiatrist Bessel van der Kolk both emphasize that the brain remains plastic across development, and that therapeutic relationships can partially remodel stress-response systems even in adults — not erase early programming, but reorganize it.

Misconception: ACEs only affect disadvantaged children. The Kaiser-Permanente sample was predominantly middle- and upper-middle class. Abuse, parental mental illness, and household dysfunction operate across income levels; the difference is that higher-income families have more resources to buffer and conceal those experiences.

Misconception: More ACEs automatically means more severe outcomes. The dose-response relationship describes population trends. An ACE score of 6 in a child with a deeply engaged, emotionally responsive caregiver may produce different neurobiological outcomes than a score of 2 in a child who is completely isolated. The presence of buffering relationships modifies the biological impact at the level of gene expression — a phenomenon documented in both animal and human epigenetics research (National Scientific Council on the Developing Child).

Misconception: ACE effects are permanent after a certain age. Neuroplasticity does not have a hard cutoff. Early intervention is more effective — documented by evidence from early intervention services for children — but the prefrontal cortex continues developing into the mid-twenties, and stress-response regulation can be improved through stable relationships, therapeutic interventions, and physical health practices throughout adolescence and into adulthood.

Checklist or steps (non-advisory)

ACE Score: What the 10 original questions cover

The following reflects the 10 categories assessed in the CDC-Kaiser ACE questionnaire. Each category counts as 1 point regardless of frequency or intensity.

1. Physical abuse — Was the child pushed, grabbed, slapped, or hit hard enough to leave marks?
2. Emotional abuse — Was the child frequently insulted, humiliated, or made to fear physical harm?
3. Sexual abuse — Did an adult or older person touch, fondle, or attempt sexual contact with the child?
4. Physical neglect — Did the child lack adequate food, clothing, housing, or medical care?
5. Emotional neglect — Did the child feel unsupported, unloved, or unimportant to family members?
6. Mother treated violently — Was the child's mother or stepmother physically abused by a partner?
7. Household substance abuse — Did a household member have a drinking problem or use street drugs?
8. Household mental illness — Was a household member depressed, mentally ill, or suicidal?
9. Parental separation or divorce — Were the child's parents ever separated or divorced?
10. Incarcerated household member — Did a household member go to prison?

Total score = number of categories in which at least one event occurred (0–10).

Reference table or matrix

ACE categories and associated long-term outcomes (population-level risk elevations)

The figures below are drawn from the original Felitti et al. (1998) study and subsequent CDC analyses. They represent odds ratios and risk multipliers compared to individuals with an ACE score of 0.

Source: Felitti et al., American Journal of Preventive Medicine, 1998; CDC ACE Data and Statistics

The 20-year life expectancy figure appears in presentations by Vincent Felitti and Robert Anda — the study's primary investigators — and is cited by the Robert Wood Johnson Foundation in its ACE summaries. It represents a statistical average across the highest-exposure group and should not be read as a forecast for any individual.

For a grounding overview of how early experience interacts with developmental trajectories across all domains, the how family works conceptual overview and the broader context at childdevelopmentauthority.com connect these findings to the wider landscape of child development science.