A Poisonous Crime Analysis
An intelligence note suggests partners poisoning their spouses is an increasing threat. Is it?
A recent Department of Homeland Security intelligence note claims that domestic partners are increasingly using poison in attempts to kill their spouses. The DHS assessment received media attention from numerous national outlets with headlines like “Use of ‘chemical and biological toxins’ to poison domestic partners has increased, DHS says.”
My interest in this was first piqued when a friend posted the below Instagram story from ABC News.
ABC’s report on the DHS intelligence note — which is not publicly available as far as I can tell — says that people are using poison more often from online black markets or made at home to poison and kill their partners. This trend has grown over the last five years and, in introducing the subject, the ABC News anchor says that “the trend is so alarming that DHS had to issue a warning.”
ABC goes on to report that the DHS intelligence note says that “Domestic partners are increasingly likely to use chemical and biological toxins…to kill or harm their partners based on reporting from the last five years.”
This story was intriguing to me as both a former CIA intelligence analyst and a current crime analyst. Are poisoning deaths — specifically from domestic partners — rising? Is this trend really alarming? Is this worth alerting large swaths of law enforcement?
Let’s dive in!
As noted, the DHS assessment is “based on reporting from the last five years.” I haven’t read the full DHS analysis since it isn’t public, but that statement stands out because — in light of the contradicting data that’ll follow — the identified trend may reflect simply more reporting on the subject rather than more actual deaths. A few high profile cases that make it into reporting can give the impression of a trend when the underlying data doesn’t support that.
The Hill also covered the DHS assessment, writing: “There were 17 documented instances of people using chemical or biological toxins against their partners between 2014 and 2025, resulting in at least 11 deaths, according to a January bulletin prepared by DHS’s Countering Weapons of Mass Destruction Office.
More than half of those incidents occurred in the past five years, the data showed.”
And Fox News reported that “DHS said in the bulletin it has “moderate confidence” that domestic partners are increasingly using chemical and biological toxins — including cyanide and ricin — to harm or kill spouses or partners”
I don’t love any analyst writing an assessment warning law enforcement about a trend of “increasingly likely” domestic poisonings based on less than 20 incidents nationwide over an 11 year period. And I have no idea how one could have “moderate confidence” in any trend involving around 10 or so incidents (more that half of 17) over 5 years.
The DHS collection of 11 reported deaths from intentional domestic poisonings is substantially smaller than the overall number of formally recorded poisoning deaths, and it’s hard to expect any sort of law enforcement action based on one death per year nationwide on average. Crime data from official sources is a far more robust collection of poisoning deaths than whatever reporting DHS is relying on.
The two best sources on poisoning murders that I’m aware of are CDC’s WONDER and the FBI’s Supplementary Homicide Report (found on the CDE). Together, these sources show that homicide poisoning deaths haven’t really changed all that much in the last five years. We don’t have full data for 2025 though we have mostly complete data from the FBI so I didn’t include the CDC figure for that year (it’s on pace for roughly the same number as 2024).
The FBI homicides by poisoning figure is much smaller than the CDC’s, but both are far larger than what DHS reported.
Homicides from poisoning have generally increased over the last 25 or so years though with 2023 having the most recorded by both the FBI and CDC — both 3 to 4 times the 2020 figure.
So the number of people being killed in poisoning homicides is higher than it was 15 years ago, but it hasn’t moved a ton in the last five years suggesting your level of risk of being poisoned to death by your spouse isn’t increasingly.
And what about the assertion that more of these homicides involve spouses as the perpetrator?
It’s a bit harder to say because CDC data doesn’t contain information on perpetrators. The FBI’s SHR data does have some perpetrator data but it’s not always great. Around 45 percent of people killed in a poisoning in the FBI data have a family relationship with the perpetrator though the SHR perpetrator information is far from complete.
Once again, 2023 was the top year for poisoning homicides by family members, but that year was more of a blip than a trend (and only accounts for 13 deaths). The FBI data shows 36 poisoning homicides between 2014 and 2025 where the perpetrator was identified specifically as a husband, wife, boyfriend, girlfriend, or ex-spouse.
The idea that we’re seeing a rise in recorded poisoning deaths remains unconfirmed in the available data. These offenses are and have always been exceptionally rare. The share of all homicides via poisoning has increased since 2000 but that rate has remained reasonably steady over the last 5 years. Around 0.8 percent of CDC homicides and 0.1 percent of FBI homicides since 2020 have been via poisoning (0.83 and 0.11 percent respectively in 2025).
Are we seeing an increase in poisoning homicides involving loved one perpetrators? The available data says no, at least not over the last five years as the DHS note assesses. Now, we don’t have data on the types of poisons being used, so it’s certainly possible that poisons like ricin and cyanide are being used more often in these cases, but there hasn’t been a measurable increase in poisoning homicides over the last few years as asserted by DHS.
These offenses are incredibly rare and noticing that more than half of 17 incidents have occurred over the last five years sounds more like noise than signal to me. They are less rare than they were 15 years ago, but that doesn’t mean the underlying analysis is accurate as a piece of intelligence, and I’m skeptical that this issue deserved an intelligence bulletin and multiple national news networks to cover it.
New on the Jeff-alytics Podcast
This week I talk with Mayor Helena Moreno about crime and criminal justice in New Orleans. We discuss the city’s difficulties, like a severe budget crisis and the ongoing challenge of abysmally low sexual assault clearance rates, as well as the city’s successes.
Mayor Moreno has had a front row seat to the city’s crime rise and subsequent decline, and she details what she attributes the historic drop to — a combination of prevention investments like expanded summer jobs programs and violence intervention, targeted policing of the most violent offenders, and critical partnerships with state and federal law enforcement. Crime and criminal justice issues in New Orleans are often a microcosm of what is happening nationally, and this conversation puts a spotlight on how to attack problems happening both in New Orleans and throughout the country.
And while you’re here, be sure to check out these other recent great episodes:
Researcher and Former Crime Analyst Carlee Ruiz
Council on Criminal Justice President Adam Gelb
Baltimore Mayor Brandon Scott
Filmmakers Ferne Pearlstein and Bob Edwards.
I wonder if the overdose crisis, particularly fentanyl, may play a role here, particularly the CDC data. LE has been somewhat more aggressive in charging the person providing or sharing the drugs resulting in death and much of that is within relationships. Texas, notably, requires fentanyl overdoses to be listed as poisonings and the medical community tends to classify overdoses as poisonings using the ICD codes. Knowing the poisons behind DHS’ conclusions would help here.
The increase from the CDC over the past ~15 years is interesting!
That trend pre-dated ChatGPT, but I think an interesting security benchmark would be whether the current GenAI tools refuse to help with something like that. (Anthropic has published about terrorism topics, but the surface of more mundane illegal stuff is clearly much broader.)