What first responder peer support programsmiss.

First responder peer support programs work well at the job they were designed for: reaching someone who has decided to talk. What they miss is everything before that decision. The program activates on self-report, and research shows self-report tends to fire once a problem has already become overwhelming. That makes peer support a response system, not an early signal.

I spent more than 20 years in the fire service, and I watched it get steadily better at offering help. Posters in the bay. A number on the back of the ID card. A trained list of people who had been where you were. That progress is real, and the people who staff these teams are doing the hardest volunteer work in the building.

This is not an argument against peer support. It's an argument about where departments file it. Most file it under prevention. It doesn't belong there, and the reason has nothing to do with the quality of the team.

What first responder peer support programs are built to do

Peer support programs train members of the workforce to provide emotional, social, and practical support to their colleagues, and to act as a bridge to professional resources. That's the design. A peer isn't a clinician and isn't meant to be. The value is that a peer clears a barrier a clinician can't: the person listening has run the same calls, so the first conversation doesn't start with an explanation of what the job is.

At that job, peer support outperforms the alternatives on the one thing departments can measure easily, which is participation. Members are more likely to walk toward a peer than toward an employee assistance line. That's a genuine result and it's why these programs spread.

Read the design statement carefully, though. Every part of it starts after contact is made.

The trigger is self-report, and self-report fires late

A peer support team has no way to know who needs it. It waits to be called, or it waits for something visible enough that somebody else calls on a member's behalf. So the real question isn't whether the team is good. It's what has to happen inside a person before the trigger pulls.

The research answer is uncomfortable. A 2020 study of firefighters and EMTs in Arkansas mapped what facilitated help-seeking, and the facilitators included recognizing that others were struggling too, encouragement from peers, previous good experiences with a clinician, and the perception that the problem had become overwhelming.

That last one is the honest description of the trigger. Not drift. Not the fourth month of sleeping badly. Overwhelming, or visible enough that other people started noticing.

Which means the months upstream of that point are invisible to the program by design. Not through neglect, not through underfunding. And the culture of this job spends 20 years teaching a member exactly what raising his hand is likely to cost him.

What the research shows about peer support, and what it doesn't

Departments deserve the honest version of this evidence, in both directions, because most of what gets published about these programs is written by people trying to sell you a training course.

The strongest positive result comes with a condition attached

The best evidence available comes from a cohort study of firefighters in Alberta, Canada, following the 2016 Fort McMurray fire. It found peer support associated with substantially lower odds of a depressive disorder (odds ratio 0.22) and an anxiety disorder (0.45). That's a real, meaningful finding, and it deserves to be quoted.

So does the condition on it. Those results held where the firefighter and the fire chief both confirmed the support existed. That agreement was present in only 45.8% of peer support reports. The PTSD result (0.62) had a confidence interval crossing 1, meaning it didn't reach statistical significance. And the study was observational, so it cannot establish that peer support caused the difference.

The authors' own conclusion is the sentence every chief should sit with: to be effective, simply offering a support is not enough, it must be recognized and acknowledged by the first responders.

The evidence base is thinner than the confidence around it

A 2025 critical review of first responder peer support programs found a shortage of randomized controlled trials or other robust quantitative methods, small convenience samples, no prevailing definition of what "peer support" even means across studies, and a lack of standardized, validated outcome measures.

That review does not conclude peer support is ineffective. It concludes we don't have the evidence base to say much with confidence, and it calls for one. Those are different statements, and the difference matters when a program is being defended in a budget meeting on the strength of results nobody has measured.

Access is not the same as a program people will use

There's a second gap, and it shows up in the numbers departments most like to report.

A survey of 2,156 professional firefighters found 81% had access to behavioral health services through their department. In the same sample, 68% said they would not recommend those services to a colleague. Most said they'd go to a spouse or family member (67%), or pay for private professional services themselves (60%), before using what the department provided.

Access was nearly universal. Endorsement was not. A program can be fully funded, fully staffed, and sitting at 81% availability while the workforce quietly routes around it.

Put that next to the 45.8% agreement figure from Alberta and a pattern shows up that has nothing to do with whether peer support is a good idea. In more than half of those departments, leadership believed a support was in place and the firefighter didn't report receiving one. That's not people lying on a survey. That's the distance between a program on an org chart and a thing that happens to a person.

Response system, not early-warning system

Here's the reframe, and it's the whole point of this piece.

Departments know their staffing to the minute and their response times to the second. Ask the same department what a specific member's physiological load has done over the last 6 weeks and there's no number, because nobody built the instrument. So the early-warning job gets quietly assigned to the only thing in the building that touches this ground at all, which is the peer team, and the peer team can only act after somebody speaks up.

Then, when it finally costs a member a marriage or a career, the review asks why he never called. He never called because the answer to "what does raising my hand cost me" has been the same for 20 years, and the department's response to that was a poster with a number on it.

What to pair first responder peer support with

Keep the team. Fund it better, train it better, and be precise with it about the job: it's your response, and a good one. Then close the part it was never built to cover.

• A signal that doesn't require volunteering. Something that reflects physiological load and recovery over weeks, on an individual baseline rather than a population average. It should surface drift without anyone having to announce it.
• Individual baselines, not population norms. Two members on the same schedule can walk off it carrying completely different loads. A department average hides exactly the person you're looking for.
• Private to the member first. If the data routes to a supervisor before it routes to the person it describes, you've rebuilt the exact career calculation that keeps hands down. Participation dies and you're back to self-report.
• Awareness, not treatment. This kind of measurement doesn't diagnose anything and shouldn't claim to. It tells a person something about their own physiology earlier than they'd otherwise notice it. What they do with that is theirs, and often the right next step is the peer team.
• An honest metric for the peer program itself. Utilization counts are not an outcome. Rising utilization can mean the program is trusted, or it can mean more people reached crisis this year. Track referrals made and connections completed, not the count of calls received.

The goal isn't to replace peer support with a dashboard. It's the opposite. Give the peer team a reason to knock on a door before that member is ready to knock on theirs.

6 questions to ask about your own program

• If we surveyed our members tomorrow, what percentage would confirm we have peer support? Not the org chart, the members.
• What is the earliest point at which our program can possibly know a member is struggling?
• What do we count as success right now, and is it an outcome or an activity?
• Of the members who used peer support last year, how many were connected to a resource, and how many of those connections completed?
• What would a member reasonably believe happens to his standing if he uses this program? Is that belief accurate?
• What number do we have that would change before a member decides to speak up?

If question 6 has no answer, that's the gap. It isn't a failure of the peer team, and it was never theirs to fill.