The Overlooked Patterns of Mental Illness Behind Violent Acts

On a recent episode of The Megyn Kelly Show, host Megyn Kelly convened Chad Ayers (founder of ProActive Response Group), Joe Leurs (former police officer and U.S. congressional candidate), and Geno Roefaro (co-founder and CEO of SaferWatch) to dissect what she called the “overlooked” warning signs that precede mass violence. 

Kelly’s premise was blunt: political fights over guns tend to eclipse the practical, pattern-based interventions that can stop attackers before they act. The panel’s throughline – drawn from school shootings to nightclub massacres – was that the danger usually reveals itself early; the system often just fails to catch it.

Geno Roefaro: “Stop the Threat” With Proactive Reporting

Geno Roefaro framed the issue in operational terms: you’re not going to prevent all access to weapons, so you must focus on the threat. He explained that SaferWatch, created after the Pulse nightclub attack and expanded after Parkland, connects tips from students, parents, and staff directly to hundreds of law-enforcement partners, documenting and routing them for action. Roefaro said the platform has helped stop 13 planned school shootings, crediting “proactive reporting” and formal accountability (documented tips, audit trails, law-enforcement follow-through) for the results.

Whether or not you use SaferWatch, this threat-management model aligns with what behavioral experts advocate: multiple reporting on-ramps, rapid triage, and interagency follow-up that doesn’t die in someone’s inbox.

Chad Ayers: Attackers Plan – And Study Other Attackers

Contrary to the “he snapped” cliché, Chad Ayers said most offenders plan for weeks or months and often try to outdo prior shooters. He pointed to manifestos, dates chosen for grim anniversaries (e.g., a would-be attacker planning for the Columbine milestone), and site familiarity (childhood church or school) as common threads. In the case discussed on air, Megyn Kelly highlighted a suspect’s hand-drawn maps of pews and an altar – what looked like meticulous pre-visualization.

This fits the research on “leakage” – when would-be attackers telegraph intent through writings, drawings, or online posts. These are moments when bystanders and institutions can intervene, if there’s a clear place to send concerns and confidence those tips will be taken seriously.

Layers of Security Help, But People Need Training

Ayers endorsed layered safeguards – key-card access, panic bars, security personnel, visitor control – but stressed they’re not enough without training and situational awareness. “The most secure home in America still has people try the fence,” he quipped. He urged basic habits: heads up, phones down, know exits, and a plan.

Physical security buys time; trained people use that time. Drills that emphasize calm, simple actions (lock, light out, low profile, communicate) can turn seconds into survivable minutes.

The Tool vs. the Threat: Weapon Substitution Is Real

On the polarizing guns debate, Ayers argued that handguns remain the most common weapon in active-shooter incidents and that banning a category of firearms won’t stop those intent on mass harm. He and Megyn Kelly cited attacks using vehicles, knives, or homemade explosives (e.g., the Boston Marathon bombers). The inference: fixate less on the tool and more on the person showing dangerous behavior.

Access matters – safe storage and timely removal can prevent impulsive violence – but Ayers’s substitution point is true in the aggregate. Policies that focus only on hardware miss other levers: threat-assessment, ERPOs (“red-flag” orders), and community reporting systems.

Cryptic Drawings, Demon Mirrors, and a Preoccupation With Killing

Kelly walked through disturbing images from the suspect’s YouTube feed: a sanctuary diagram with pews and altar; coded letter/number strings; a self-portrait staring into a mirror that reflects a horned demon; and ammunition marked with phrases she described as “for the kids” and “humanity is overrated.” In Kelly’s telling, the content read like the notebook of other infamous shooters – obsessive planning, grandiosity, and dehumanization of targets.

Content analysis can’t predict violence with certainty, but dehumanizing language, fixation on prior attackers, and articulated desire to kill are high-risk flags in threat-assessment models. Those flags must be reportable, reviewable, and actionable.

“They Need to Be Locked Up”: The Hard Claim, and the Harder Balance

Kelly characterized many perpetrators as in “deep depression and madness,” saying they “can’t be persuaded… can’t actually be therapized… [and] need to be locked up.” It’s a raw, understandable reaction to catastrophic harm. Still, blanket confinement based on mental-health labels raises civil-liberties concerns and misstates risk: most people with mental illness are not violent, and most violence is not caused by serious mental illness.

We should be precise. Imminent risk (explicit threats, weapons acquisition, plans) warrants immediate action – temporary removal of access, involuntary evaluation when legal thresholds are met, and court-supervised conditions. But broad calls to “lock them up” can backfire by scaring families away from seeking help.

Roefaro: “About 90%” Show Warning Signs – So Why Aren’t They Reported?

Citing case patterns, Geno Roefaro said “about 90%” of perpetrators display known warning signs – leakage to friends, social media posts, “hit lists,” stockpiling gear. He pointed to Pulse, where, he said, a coworker noticed weapon and ammo purchases but never reported concerns. Roefaro’s critique of “call 911” slogans was practical: people rarely dial 911 for incipient threats; they need low-friction, semi-anonymous ways to share concerns that still reach the right investigators.

This is the linchpin. Systems that normalize early, non-emergency reporting (apps, web forms, dedicated lines) and guarantee feedback to the reporter can surface risks before the legal threshold for arrest is met.

Joe Leurs: “Call It What It Is – Mental Illness,” and Focus on the Person

Joe Leurs argued that euphemisms obscure the problem: we should name mental illness when we see it, follow “if you see something, say something,” and scrutinize how the gun was obtained. He criticized overreliance on medication – “no pill is going to fix that” – and emphasized offender patterns like choosing soft targets (schools, churches) over hardened sites (police stations, military bases). He also applauded Kelly for not naming perpetrators to deny notoriety.

Focus on people, yes – but with nuance. Mental-health symptoms alone do not predict violence; specific behaviors do: threats, interest in prior attackers, weapon acquisition, time/place scouting, and leakage. Precision keeps us from stigmatizing millions who need care.

The Transgender Flashpoint: Identity, Treatment, and Violence

Kelly raised a fraught question: recent high-profile cases where shooters were transgender or nonbinary. She criticized what she called a “capture” of the medical system that “affirms, affirms, affirms” instead of probing underlying mental-health issues. Chad Ayers agreed that “everyone who commits one of these attacks is in some type of mental crisis,” and opposed “coddling.” Kelly referenced several cases (Nashville, Colorado Springs, others) to argue the topic can’t be off-limits.

I feel like it’s legitimate to discuss co-occurring mental-health challenges and to demand rigorous, ethical standards of care. But we should not imply a causal link between gender identity and violence. The overwhelming majority of transgender people are not violent, and many are themselves at higher risk of victimization. Preventive strategies should focus on behaviors and leakage, not identity categories. We can debate clinical models without scapegoating a group.

Should Media Name Shooters? Kelly Says No – And the Research Backs Caution

Megyn Kelly reiterated her policy: do not name the perpetrator; center victims. She cited expert criticism (including Gavin de Becker’s advocacy) that naming and glamorizing offenders feeds contagion – the copycat effect Ayers mentioned. The clip Kelly’s team played – of the suspect knifing his own notebook while muttering “kill myself” – underscored the blend of self-hatred and fame-seeking that can drive these attacks.

The “No Notoriety” approach is sensible: report essential facts, deny the branding, and focus on patterns and prevention.

The Bottom Line: Patterns Are Visible – If We Choose to See Them

Megyn Kelly pushed the conversation toward warning signs and patterns. Chad Ayers underscored the premeditation and the necessity of training. Geno Roefaro argued for documented reporting pipelines that convert tips into action and claimed concrete prevention wins. Joe Leurs urged straightforward language about mental illness and a relentless focus on people over tools. My takeaway: we can reject scapegoats and still act decisively. Look for behavioral leakage, build frictionless reporting, train humans, and ensure lawful, proportionate interventions. That’s not left or right; that’s how you stop the next tragedy before the first shot is fired.