Decommissioning the Human Tracking System
For the better part of two years, I was the medical tracking system for my organization. Not by design. Just by default.
It was just personnel texting me when something was wrong, sometimes with a photo of a doctor’s note, sometimes with nothing at all. Me making clinical decisions without the records I’m required to have, often without ever having met these people. Me absorbing the friction so the machine could keep moving.
The problem with being a load-bearing workaround is that it works. It’s uncomfortable in the way carrying something heavy is uncomfortable. You can do it, but you’re aware of the weight the whole time, and you’re wondering why all this crap has to be in your backpack.
So I wrote it down.
What the system actually looked like
Some context: personnel in this organization don’t have the structure of a typical organization. We see each other infrequently, and our job demands peak physical condition. Medical issues happen in the interim, and when something comes up, the path of least resistance is to contact the unit’s medical clinician to get a note providing duty restrictions.
What I said: I need actual records before I can make any determination about your fitness for duty.
What I meant: I cannot legally or ethically tell you whether you can do the work based on a text message.
What happened: I got photos of “return to work” slips. Voicemails. Texts that said “doc said I can’t do a fitness test” with no further context. Occasionally nothing at all, and I’d find out weeks later that someone had been restricting their own activity based on guidance that was never reviewed through the proper channels and may not have applied to their job at all.
Our leadership wasn’t in the loop because there was no loop. There was just me, and I’d run out of bandwidth juggling texts, emails, spreadsheets, and finicky medical readiness systems.
The design failure
Personnel are doing what makes sense given the options available. If texting the clinician gets a faster answer than navigating a process that doesn’t clearly exist, they’re going to send that text. That’s rational.
The workaround persists because it works for most people in the short term. Personnel get informal guidance. I get enough information to make a reasonable call. Nobody has to navigate a process that was never built. The only person absorbing the cost is me, and I’ve been doing it quietly most of the time, so there’s no visible crisis.
Which is to say: the system isn’t broken. It’s just optimized for something I’m not sure we want to be optimizing for.
What I built
A formal policy document laying out a tiered classification system: minor concerns that need no paperwork, routine injuries requiring documentation and a duty restriction, complex cases requiring formal tracking and a full investigation into whether the injury happened on duty. A notification template that tells supervisors exactly what I need and when. A counseling form template with language that matches the policy, so there’s no ambiguity about what someone was told or when they were told it.
The documentation itself took less time than I expected. The harder part was framing the email to the staff.
Because here’s the thing about building infrastructure when enforcement lives in someone else’s lane: I can design the best process in the world and it changes nothing if it’s not used. So I put it plainly. If we define a standard and then accept non-compliance without consequence, we’ve just created paperwork, not a standard. If personnel are ignoring documentation requests for months while claiming they can’t do their jobs, there’s a bigger problem than their recovery. And if there are no real consequences for someone whose medical status is genuinely unknown to everyone responsible for their safety, this system is theater. I have very little interest in participating in that.
This isn’t a new problem. Every previous attempt to formalize this process has stalled in the same place: someone wants to get the enforcement policy exactly right before anything gets signed, the review drags as we wait on subject matter experts, and eventually we’re back to texting. I understand the impulse. But we’ve run that play. I cited the relevant regulations and the specific authority leaders already have. What I’m asking for isn’t new policy. It’s acknowledgment that the tools exist, and a willingness to use them.
Where this goes from here
This won’t be my system to carry anymore, but I want it fixed before I hand it off. We have a new admin for tracking communication flow, and an operations officer to provide oversight. Two positions that have been effectively empty for a year. Both are new to their roles, which is actually the right moment, before habits calcify around whatever informal process was working before. The policy is an attempt to build something they can own without needing to reverse-engineer what I was doing in my head.
If the staff comes back with feedback, I’ll revise. If leadership identifies what enforcement looks like in practice, I’ll incorporate it. I can control the documentation I generate. I can control what happens in the room with each patient. I can control the system I design and operate. That’s it. Not the command climate. Not whether anyone reads the email. Just the work I put in front of them.
It’s not the most satisfying place to land. But I’ve been carrying this long enough, lightening it just a little bit before the handoff still feels like a win.