If you or someone else is in immediate danger, call 911

Child Abuse Reports in the U.S.: What the Numbers Show—and What They Miss

Explore the importance of confidential tips in the reporting process and how they build trust with victims.

When people picture child abuse, they often imagine a “bad stranger.” The reality is usually the opposite: abuse and exploitation most often happen in relationships where the child has dependence, trust, and an authority/power imbalance—inside families, institutions, and communities.

That authority dynamic matters, because it affects whether a child (or a now-adult survivor) ever feels safe enough to tell—and whether anyone believes them when they do.

What gets counted in “official” numbers

Most national statistics come from child protective services (CPS) agencies reporting into the National Child Abuse and Neglect Data System (NCANDS). These numbers are best understood as maltreatment known to CPS, not “total abuse that happened.”

In federal fiscal year 2023, the Children’s Bureau reported that CPS agencies responded to allegations involving about 3.1 million children, and 546,159 children were found to be victims of abuse and/or neglect.

The same national report also documented about 2,000 child maltreatment deaths in 2023.

And in terms of the type of maltreatment CPS substantiates/indicates, neglect is the most common—the Children’s Bureau reported 64% of victims experienced neglect (children can have more than one maltreatment type in a case).

A Tennessee snapshot (example of state-level reporting)

Federal NCANDS-based state pages show how widely the “known to CPS” numbers vary by state. For Tennessee in 2023, the federal outcomes site reports:

Tennessee’s own Department of Children’s Services (DCS) annual reporting also provides context on intake volume. For example, its FY2024 annual report lists 93,743 hotline calls handled and 77,003 web reports/referrals (note: these are operational intake measures and don’t equal “substantiated victims”).

Who are the perpetrators—and where do “authority figures” fit?

National CPS data commonly groups perpetrators by relationship categories rather than job titles (for example, it doesn’t usually say “pastor” or “teacher” as a standard national category).

One commonly cited breakdown (using national child maltreatment data) shows that in 2023 alleged/substantiated perpetrators were most often:

That “other nonparent” category can include many possibilities—neighbors, family friends, babysitters, coaches, clergy, teachers, and others—so it’s evidence that non-parent authority/trust relationships matter, but it does not provide a clean “how many pastors/teachers” count.

How much abuse goes unreported?

There’s no single “perfect” number, because “reported” can mean different things:

But multiple research streams point the same direction: a large share of abuse—especially sexual abuse—is never reported to authorities.

For example:

Also important: the same UNH report notes reporting is less likely when assaults occur at school, and it discusses how “other officials” (like schools) sometimes become gatekeepers or alternatives to police involvement.

Why abuse by authority figures is especially hard to disclose

When the abuser is someone “in authority” (a pastor, teacher, coach, parent, step-parent, or household authority), disclosure can be blocked by:

A major research update on child sexual abuse disclosure identifies power dynamics and social-context barriers as key reasons disclosures are delayed or never happen.

A public inquiry summary on barriers to disclosure also explicitly names the perpetrator’s position and authority as a factor that can suppress disclosure (along with grooming, threats, isolation, and shame).

Can we show “how many pastors” or “how many teachers” offend each year in each state?

Not responsibly—at least not with a single national, authoritative source.

There is no comprehensive, standardized U.S. dataset that reliably reports annual, state-by-state counts of child abuse perpetrators specifically labeled as:

CPS and law enforcement data tend to classify perpetrators by relationship (parent, relative, nonparent) and case context, not by profession. Some jurisdictions may track “educator misconduct” or clergy cases in specific systems, but those aren’t unified nationally.

What you can do on your site (and stay accurate):