The risk in a return-to-office order isn't who walks out. It's what happens to the ones who stay.

You've got the return-to-office memo drafted and sitting in a folder called "Q4 — hold." Four days in the building, starting in October. Your operations lead has been asking for it since spring — she says the culture is thinning, that nobody knows what anyone else is actually working on anymore. Two of your best account people moved out of state a few years back and built their whole lives around not commuting. You don't fully know which fight you're picking: the one about culture, or the one about control.

A study published in July in the journal Frontiers in Psychology looked at 7,704 employees at one large healthcare system in the southwestern United States, split into fully remote (1,869), hybrid (2,099), and fully onsite (3,736) groups. The researchers — Alyssa Lezcano, Stephanie Zajac and Courtney Holladay of the MD Anderson Cancer Center Leadership Institute, with Stefanie Johnson of the University of Colorado Boulder's Leeds School of Business — combined an employee well-being survey from 2023 with the organization's own turnover records from a year later. It's worth naming the limits plainly: this is one organization, in one industry, where remote roles clustered in administrative and analytic work and onsite roles clustered in clinical and operational work, so location and job type aren't cleanly separable. Well-being was self-reported on a standard scale. Connection to coworkers was inferred from open-ended survey answers run through language-analysis software, not observed directly.

The headline finding isn't surprising: remote employees reported meaningfully higher well-being than hybrid employees, who reported higher well-being than fully onsite employees. What's less expected is what didn't move. Turnover itself, measured a year out, barely differed by work location — 7.8 percent for remote, 8.3 percent for hybrid, 8.8 percent for onsite, a gap the researchers found was not statistically significant on its own. Location didn't directly predict who left. Well-being did. Work environment only shaped turnover indirectly, by shaping well-being first.

"Taking away people's choice of how they work is probably not going to help them in terms of their well-being."

Stefanie K. Johnson, Leeds School of Business, University of Colorado Boulder — as quoted in CU Boulder Today's coverage of the study

Here is why this lands differently for you than for a department head at a bigger company. A large employer has an HR function tracking engagement scores on a quarterly cadence, flagging drift before it becomes a resignation letter. You don't. You find out your best person is checking out when she stops raising her hand in the Tuesday meeting, or when the numbers on the account she owns start slipping and you can't tell whether it's the market or her. In a company like yours, well-being isn't a line on someone else's dashboard — it's the actual difference between a person still solving problems for you and one who's merely present for them. The study's finding that turnover lagged well-being by a full year is, for you, the real warning: by the time the resignation shows up, whatever caused it has usually been sitting there for months, unmeasured, because you had no instrument pointed at it.

The part the research doesn't say, but that I'll say plainly: this study can't tell you whether your particular return-to-office push is about culture or about not trusting your people out of your sight. Those are different problems, and a four-day mandate solves neither if it's aimed at the wrong one. If the real issue is that nobody knows what anyone else is working on, that's a visibility and accountability problem — you can build that without requiring five days of commuting to get it. If the real issue is that you don't believe the work is happening unless you can see it, no location policy fixes that. It just moves the mistrust into the building along with everyone else.

One thing to do this week, before you finalize anything: ask each of your team leads one question in writing. Not "how's the team doing." Ask: "Name one person you're not fully sure is still in it, and tell me why." You'll get a shorter, more honest list than any survey would produce, and it will tell you whether your actual problem is where people sit or something an office was never going to fix.

Whether you reach for a mandate or a workaround here usually says less about your industry than about how you're wired to handle the loss of direct oversight. If you want a clearer read on that before you send the memo, the Leadership Profile is built for exactly this kind of decision — take it at /wiring-map.