Heat Has No Off-Season: The Indoor Workplaces Where Summer Never Ends

Indoor Heat Stress Is Not Seasonal

In Birmingham, Alabama, in the third week of October, the crew walking into the iron foundry has jackets on. Forty minutes later two of them are on the pour floor in aluminized gear, standing eight feet from a ladle of molten iron, and nothing on that floor is a degree cooler than it was in August. The weather changed. The furnace did not.

The question worth asking in that building is not whether it’s hot outside. It’s whether the heat program that ran in July is still in place, and whether anyone is still measuring. In most plants the answer to both is no, because heat programs are written against the season rather than against the conditions.

That’s a reasonable mistake and an expensive one. The heat a foundry, a bakery or a server hall imposes on a body is manufactured on site, continuously, by equipment that doesn’t read the forecast.

Is Heat Stress Only a Summer Problem?

The Indoor Half of the Data

The clearest federal picture of where heat illness actually occurs comes from 66 heat-related illness investigations handled between 2011 and 2016, analyzed by Tustin and colleagues in the CDC’s Morbidity and Mortality Weekly Report and carried into OSHA’s own background document for the heat rulemaking. Two findings from that set matter, and they point in opposite directions.

The first: 61% of the nonfatal cases occurred during or after work in an indoor environment. The second, from the same investigations: 80% of the fatalities were outdoors. Read together, they describe a real division of labor between the two settings. Outdoor work, in open sun with no ceiling on exposure, is where heat kills. Indoor work, where the heat source is fixed and the shift is structured, is where heat quietly degrades people, sends them home early, and generates the recordable heat incidents.

That second category is the one a plant actually pays for week to week, and it’s the one a summer-only program misses entirely. We have put numbers to that bill before; the relevant point here is that the cost arrives in October too.

Five Indoor Workplaces OSHA Already Names

When OSHA opened its heat rulemaking it published a list of indoor industries where heat exposure is a known hazard. It’s not a short list, and none of these work environments get cooler in winter:

The Workplace Missing From the List

The list was written before the current massive build-out of AI compute capacity. A contained hot aisle in a modern data hall runs well above the temperature of the cold aisle a few feet away, because the heat a rack rejects is set by the hardware rather than by the building, and the people who work rear-of-rack stand in the exhaust. Nothing about that is seasonal. Economizers change where a hall draws its intake air in winter; they do not change the heat the servers make.

That environment is not on OSHA’s indoor list yet. It’s covered by the general responsibility to provide a safe workplace, and it’s inside the 55 industries the renewed emphasis program directs inspectors toward, but it has no identified place in the agency’s own description of indoor heat work. That gap is worth watching, and it is the subject we take up later in this series.

Why Cooler Weather Does Not Reset the Risk

There is a second mechanism in play once outdoor temperatures drop, and it works against the crew. Heat tolerance is an adaptation, not a trait: it’s built by repeated exposure and it fades without it. A crew that spent August acclimatized to indoor and outdoor heat together loses part of that adaptation through a cool autumn, then walks back onto the same pour floor with less margin than it had in summer.

The OSHA case series makes the stakes clear. Of the 14 fatal illnesses in that set, 11 involved workers the authors classified as unacclimatized, which they defined as having started within the previous two weeks or returned from more than a week away. Holiday shutdowns, seasonal hiring and January restarts all produce exactly that worker, in a building whose heat never changed.

What Keeps Indoor Crews Safe Twelve Months a Year?

Measure On Site, Not From the Forecast

OSHA is unusually direct about this. Its hazard-recognition guidance states that “a weather report cannot gauge conditions inside a building” and that “workplace environmental heat should be measured on-site using WBGT meters.” Those two sentences quietly invalidate most indoor heat programs, which trigger on a regional heat index and therefore go dormant the week the weather turns.

On-site measurement also surfaces the thing a forecast structurally cannot: the gap between the room and the task. Two workers in the same bay, one on a forklift and one swinging a sledge in the same gear, are carrying different loads, and the threshold the agencies converge on is a property of the body rather than of the room.

What a Year-Round Program Contains

What a Program Catches Before It Becomes a Recordable Incident

The American Foundry Society’s health and safety committee surveyed how member foundries control heat exposure and published the results in 2020. The most useful figure in that report is not an injury count. Across eight foundries with established programs, the committee recorded more than 100 early-warning incidents in three years: the cramping, the dizziness, the worker pulled off the floor and cooled before anything went on a log.

That’s what a working program produces. It doesn’t eliminate heat strain, because the furnace is still there. It converts strain into something a supervisor sees early, in a building where the alternative is finding out too late – when a worker goes down. A program that generates a hundred small catches and few recordables is functioning exactly as designed, and a program that generates neither is usually not measuring.

When This Does Not Apply

Indoor heat is not a universal hazard, and treating every warm room as a heat-stress problem costs credibility with the people who have to implement the fix. This is the wrong conversation in these cases:

The One Control That Travels

Controlling the environment comes first and always will: ventilation, shielding, enclosure, remote operation, process change. Where heat can be removed, it should be. The problem in most of the industries above is that the heat is a necessary byproduct. A foundry cannot stop melting iron, a kitchen cannot stop running ovens, and a data hall cannot stop rejecting what the racks produce.

What is left is the person and the interval. The body already defends its core temperature by moving blood to the surfaces that can shed heat, and that system is the one worth assisting rather than overriding. Cooling the palm during a rest break pulls heat out through the body’s natural radiator, which is why the method works the same way in a January boiler room as it does on an August job site, and why the Army settled on cooling the body between work bouts rather than cooling the air around it.

It also travels where a cooling station cannot. That matters most in the places where the gear itself is the insulation, because the break is the only moment the gloves come off. The test of whether any of it is working is not comfort: it is whether the same task produces a lower heart rate at the same point in the next cycle.

Key Takeaways

External References