Why protecting Colorado children from dying of domestic violence is such a hard problem

A record number of Colorado children^[1] died in 2024 as a result of domestic violence, despite a statewide reduction in overall homicide.

Of the eight children who died, five were involved in active custody disputes. These deaths took place when families faced high stress but also when legal systems should have been well placed to intervene. Multiple children were killed alongside a sibling or a parent^[2].

As a researcher studying domestic violence, crime and anti-violence policy^[3], I have watched these numbers with a sense of resignation rather than surprise.

Domestic violence homicide is persistent. Local, state and federal governments spend millions of dollars each year^[4] to operate hotlines, fund shelters and engage in prevention programs for victims of domestic violence. Yet more than one-third of homicides of women^[5] are still perpetrated by intimate partners. And there has been a steady increase nationally in domestic violence-related deaths of children^[6] over the past 20 years.

It’s clear that something is different about domestic violence that resists our attempts to reduce overall violent crime. But researchers have struggled to identify exactly what those differences are in ways that can inform effective policy.

To start addressing these deaths, we first need to effectively measure them, a task that is more challenging than one might expect.

Measuring domestic violence

Studying domestic violence is, at best, difficult — not least because data is highly limited.

Researchers often try to ask causal questions about what works to prevent domestic violence. To do this, they use large-scale national datasets, including the Uniform Crime Reporting Program^[7] and the National Incident-Based Reporting System^[8]. However, these datasets are often incomplete or have inconsistent reporting from responding agencies^[9].

Law enforcement may not recognize and interpret a fatality as resulting from domestic violence if abuse was not previously reported. It is particularly challenging to identify whether a death involved dating or sexual partners unless witnesses who knew the victim closely cooperate with the investigation.

Additionally, the vast majority of victims of domestic abuse do not contact law enforcement^[10] or seek medical care^[11]. Often, this is due to fears that police will not believe them or that their abuser will find out. Parents may worry their abuser could take custody of their children, or that calling 911 will instigate child welfare system involvement.

The result is that half of the perpetrators of domestic violence fatalities in Colorado in 2024 did not have a prior domestic violence-related arrest^[12]. Only one-fifth had been previously convicted of domestic violence.

Domestic violence affects more than intimate partners

Domestic violence affects more than intimate partners or spouses^[13]. It can also affect siblings, roommates and even neighbors, co-workers or bystanders. These are collateral victims – people harmed by domestic violence without directly being part of the abusive relationship.

Colorado^[14] and Wisconsin^[15] have expanded their definition of domestic violence fatalities to account for some of these collateral deaths. For years, Colorado has included abusers who died by suicide^[16], or whom law enforcement killed in the line of duty, in statewide counts. But states disagree on how wide to cast the net, making comparisons between states difficult.

These fatality reviews are further hamstrung by the boundary between domestic violence and child abuse.

In Colorado, deaths due to child abuse and neglect are counted in the Domestic Violence Fatality Report^[17] only if the death can be traced to violence between intimate partners^[18]. Children can therefore get lost in the count when violence between parents or caregivers is hidden behind closed doors.

What we don’t know can hurt us

These data gaps present challenges to understanding^[19], predicting^[20] and preventing^[21] domestic violence. Policymakers struggle to gather up-to-date information to make effective public safety policy, including over how and when to detain alleged abusers before their day in court.

In Colorado, pretrial detention recommendations are made using a rigid scoring rubric^[22]. This rubric includes the accused’s prior criminal sentences or time served in jail or prison. However, it does not include information about domestic violence protection orders or prior charges that did not result in conviction.

In general, this is a well-intended policy that upholds the principle of “innocent until proven guilty.” But in domestic violence cases, it creates a catch-22. The vast majority of abusers have never been found guilty in court. This can be due to dropped charges^[23], lack of victim cooperation or unclear evidence. These abusers can have long histories of abusive behavior that aren’t visible to a judge when making pretrial detention decisions.

Designing effective prevention and response

Despite these challenges, policymakers have made substantial steps forward.

In 2022, the national Bipartisan Safer Communities Act^[24] closed the so-called “boyfriend loophole^[25]” whereby married individuals convicted of domestic violence offenses were prohibited from gun ownership but dating partners were exempt. This is particularly important given that the majority of firearm mass shootings^[26] in the U.S. are domestic violence-related.

States and counties nationally are improving the way courts assign pretrial detention^[27] and arrest and charge offenders. Mandatory arrest policies^[28] require law enforcement to make an arrest when they suspect abuse. No-drop orders^[29] prevent abusers from intimidating survivors into dropping charges.

However, these laws have limited effectiveness^[30] and introduce new harms, including increasing domestic violence homicides^[31]. Colorado’s own mandatory arrest law has been criticized^[32] for increasing arrests of victims of domestic violence. This can threaten victims’ own custody^[33] of their children and cause further economic precarity, increasing the risk of lethal violence.

Because laws and law enforcement cannot do everything or support every survivor, solutions must come from outside of the criminal-legal system. Community-based services and programs^[34] such as emergency housing, counseling and cash assistance help survivors to overcome barriers to safety.

However, access to these programs and services varies. Not all counties – in Colorado or most other states – have emergency domestic violence shelters. Recent federal funding cuts^[35] threaten many programs’ continued operations. Even when programs exist, local availability of housing and services can limit service providers’ effectiveness^[36] for their adult clients and their children.

Failing to effectively measure, prevent and respond to domestic violence can be a matter of life and death. Given how survivors’ needs vary^[37], policymakers need to recognize that policy solutions and programs are not one-size-fits-all. And tailored, local policy solutions require improved data and better resources.

Read more of our stories about Colorado^[38].

References

1. ^{^} record number of Colorado children (www.denverpost.com)
2. ^{^} alongside a sibling or a parent (coag.gov)
3. ^{^} researcher studying domestic violence, crime and anti-violence policy (sites.google.com)
4. ^{^} spend millions of dollars each year (nnedv.org)
5. ^{^} more than one-third of homicides of women (bjs.ojp.gov)
6. ^{^} steady increase nationally in domestic violence-related deaths of children (doi.org)
7. ^{^} Uniform Crime Reporting Program (www.fbi.gov)
8. ^{^} National Incident-Based Reporting System (cde.ucr.cjis.gov)
9. ^{^} often incomplete or have inconsistent reporting from responding agencies (doi.org)
10. ^{^} do not contact law enforcement (www.thehotline.org)
11. ^{^} seek medical care (www.axios.com)
12. ^{^} did not have a prior domestic violence-related arrest (coag.gov)
13. ^{^} more than intimate partners or spouses (www.womenagainstabuse.org)
14. ^{^} Colorado (violencefreecolorado.org)
15. ^{^} Wisconsin (www.endabusewi.org)
16. ^{^} abusers who died by suicide (coag.gov)
17. ^{^} Domestic Violence Fatality Report (coag.gov)
18. ^{^} traced to violence between intimate partners (coag.gov)
19. ^{^} understanding (doi.org)
20. ^{^} predicting (doi.org)
21. ^{^} preventing (doi.org)
22. ^{^} rigid scoring rubric (capscolorado.org)
23. ^{^} dropped charges (www.news5cleveland.com)
24. ^{^} Bipartisan Safer Communities Act (www.congress.gov)
25. ^{^} boyfriend loophole (cebcp.org)
26. ^{^} majority of firearm mass shootings (doi.org)
27. ^{^} way courts assign pretrial detention (loyolaccj.org)
28. ^{^} Mandatory arrest policies (coloradonewsline.com)
29. ^{^} No-drop orders (doi.org)
30. ^{^} limited effectiveness (doi.org)
31. ^{^} increasing domestic violence homicides (doi.org)
32. ^{^} has been criticized (coloradonewsline.com)
33. ^{^} threaten victims’ own custody (www.ucpress.edu)
34. ^{^} Community-based services and programs (nnedv.org)
35. ^{^} Recent federal funding cuts (time.com)
36. ^{^} limit service providers’ effectiveness (doi.org)
37. ^{^} survivors’ needs vary (moodle2.units.it)
38. ^{^} Colorado (theconversation.com)