Healthy eating when working from home: the fridge is five metres away

What actually changes when you work from home

There is less research than you would imagine, and it pays to be precise about what it says. The most useful study is a survey of more than 33,000 Japanese workers run during the pandemic (Kubo and colleagues, 2021) that analysed 13,468 office workers who telecommuted. Those telecommuting more than four days a week more often reported skipping breakfast, eating alone, having fewer meals a day and replacing meals, compared with those who almost never did. Three honest caveats: it is a CROSS-SECTIONAL study — it measures association, not causation, so it does not say remote work makes you eat worse — it measured habits, not calories or weight, and it was run in 2020, under a forced kind of remote work that isn't quite today's. What it does sketch is a recognisable pattern: the day loses its structure.

Invisible snacking: the food that doesn't feel like eating

The coffee with the biscuit that comes alongside. Three nuts from the jar on top of the fridge. Two spoonfuls of yesterday's leftovers, standing up, while something else heats. The piece of bread while you cook. None of it feels like a meal, and that's exactly why it almost never gets recorded — not in your head, not in any app. There is no study putting a number on what it adds up to across a work-from-home day, and I'm not going to invent one; what we do know is that what isn't recorded doesn't make it into the count, and at home there are far more chances for that to happen than in an office.

Distance is the best-studied lever

This is where the experimental evidence sits, though its certainty is low. A Cochrane review (Hollands and colleagues, 2019) gathered 24 randomised trials on food availability and proximity — 18 of them on proximity: placing a food FURTHER AWAY may moderately reduce how much of it is eaten, and the effect appears to grow with the distance. Two caveats the reviewers themselves stress and that can't be hidden: the certainty of the evidence is low to very low, and over half the trials (14 of 24) were run in the lab, not in real kitchens. With that caution, the signal is consistent. In the more solid of Hunter and colleagues' two laboratory trials (2018) — the second, with 246 people — moving a bowl of snacks from 20 cm to 70 cm was enough for fewer people to take any: 57.7% versus 70.7%. In the first, with 159 people, the difference pointed the same way but did not reach significance: 53.8% versus 63.3%. And a third study (Knowles and colleagues, 2019) added the other lever: EFFORT. In that trial, wrapped brownies were eaten less than unwrapped ones, on top of the distance effect.

A 15-minute kitchen redesign

Translated into things you can do today, before the next video call. This isn't a diet: it's moving objects so the impulse meets friction and the good option doesn't.

• First thing you see when the fridge opens: washed and cut fruit, yogurt, some protein ready to go. Let it win the visual race.
• Snacks out of sight and wrapped: high cupboard, closed bag, opaque jar. In that trial, wrapped got eaten less than unwrapped.
• No food on the desk. Not even the «healthy» nut jar: distance is the lever, not the food label.
• Prepare what you want to win. One hour of batch cooking on Sunday decides more days than any resolution.
• Keep something genuinely filling within reach for mid-afternoon, which is the low point for a lot of people.

If-then hours (and why they aren't enough alone)

The second layer is your schedule, written as «if X happens, then I do Y». A meta-analysis of 70 interventions with nearly 9,700 people (Carrero and colleagues, 2019) found these plans work — but with an asymmetry worth knowing before you hand them the whole problem: the effect is small-to-moderate (d = 0.33) for INSTALLING healthy behaviours and small (d = 0.18) for REDUCING unhealthy ones. Translated: an if-then plan is good at making the 11 am fruit happen, and weak at making the urge to open the cupboard disappear. That is why this article runs environment first, schedule second. Concrete examples: «if a meeting ends, then I drink water and step outside for two minutes»; «if it's 11 am, then I have a real breakfast in the kitchen, not in front of the laptop»; «if I've finished lunch, then I wash up and close the kitchen».

Breakfast for people who get up ten minutes before logging on

Skipping breakfast did show up in Kubo's study, though it deserves to be told in full: it was the WEAKEST of its four associations (OR 1.15) and the only one without a clear dose-response trend — the strongest by far was having fewer meals a day (OR 2.39). It still makes complete sense: with no commute, the morning starts when the laptop opens. The problem isn't skipping it on principle — some people do fine without breakfast — it's skipping it without deciding to and arriving at noon with a hunger that chooses for you. If you do eat, make it genuinely high in protein: here are the breakfasts that hold a morning best and here is the daily amount that makes sense. And if you decide not to, make it a decision, with a planned meal waiting.

Office days and home days: two templates, not one

The mistake almost everyone on a hybrid schedule makes is trying to hold the same plan across all five days. They don't work the same: the office day has a commute, a collective break and lunch either solved or bought; the home day has a kitchen, silence and no barriers at all. Build two different templates and stop feeling inconsistent. On the home day, the lever is the environment and closing the kitchen. On the office day, the lunchbox and the vending machine are a different article. And on both, the steps: working from home deletes the commute at a stroke; it is the most obvious change of the day, though none of this article's sources measured how much it weighs.

Logging the snacking without it becoming a punishment

If the core problem is that some food doesn't feel like food, the answer isn't to promise yourself you'll eat less: it's to see it. And friction is the enemy again — nobody is going to search a food database for «plain biscuit» at 11:40 between two meetings. That's why photo logging changes the maths: when logging costs three seconds, the small stuff gets logged too, which is exactly what was escaping. No guilt, no drama: the goal isn't a perfect day, it's a VISIBLE one. The rationale for why this works is in why a food diary works, and if what actually happens is that you eat from anxiety rather than hunger, that has its own guide in emotional hunger versus real hunger.