At 2:15 a.m. you press a palm into the mattress like you’re checking a loaf of bread. Firm? Soft? Is that a dip, or have you just spent nine years lying in the same spot? You roll over, which is worse, and start building a case against the thing. By 2:30 you’ve priced a replacement.
Everyone’s been there, and it gets expensive, because the honest answer to the question underneath it is more awkward than anything a mattress company wants to tell you. So here’s the awkward version: what the research says about whether a new mattress improves sleep, who paid for a good chunk of it, and how to find out for yourself before you spend real money.
No brands in here, and nothing I’d tell you to go and buy.
Who’s answering your question
Search a few mattress questions and look at who shows up. When I did, most of it was companies that sell mattresses, plus trial-policy pages and buying guides. Not a scandal, just useful to know before you swallow the confident lines, like “medium-firm is best” or “replace it every eight years.” The people saying them have somewhere to be at the end of the sentence.
What you’d want instead is studies paid for by someone who doesn’t profit, using real sleep measures, with a comparison group. Those are scarce. One review I’ll get to found only seven randomized trials among 39 papers, and what they say is a lot quieter than the sales pages.
Does a new mattress improve sleep? The study everyone cites
The one you’ll meet most often is a 2009 study from Oklahoma State, in the Journal of Chiropractic Medicine [1]. Bert Jacobson and colleagues recruited 59 adults who reported some disturbed sleep and minor discomfort, with beds averaging about nine and a half years old. For 28 days they rated their sleep every morning on their own bed. Then everyone got a new medium-firm one and rated another 28 days. Sleep scores went up, and kept going up across the four weeks.
That’s the headline. The rest is in the paper too, to its credit. No control group, so nobody was compared against a new bed that wasn’t supposed to be better, or against no new bed at all. Everyone knew the new bed was free, and the authors themselves say novelty and simply being in a study could explain part of the result. Everything was self-rated, nothing objective. A convenience sample, four weeks, and funded by the Better Sleep Council, a bedding industry group.
I’m not saying it’s wrong. A new bed may well have helped. I’m saying the design can’t tell “the bed is better” apart from “people given a free bed and asked about their sleep every morning will report sleeping better.” That’s the whole question, and it’s the one this study can’t answer.
The review that tried to sort it out
A 2021 systematic review by Caggiari and colleagues pulled together 39 papers on how mattress type relates to sleep quality and comfort [2]. Its conclusion: medium-firm mattresses tend to promote comfort and sleep quality.
Look at what holds that up, though. Most of the studies weren’t randomized, and the authors rate the overall strength as moderate-to-low. Firmness was mostly judged by feel, and only one study used the European standard firmness scale, so “medium-firm” in most of this research is closer to an impression than a measurement. The results weren’t unanimous either. One study found no significant link between firmness and sleep quality, and suggested softer might even be deeper for some people. Another found a latex mattress improved objective sleep measures but not how people rated their sleep.
The sentence I’d pin to the wall: the authors say manufacturers’ claims about therapeutic or “orthopaedic” mattresses aren’t well supported by the evidence. (I’ve read the abstract and a summary, not the full review, so double-check the details before quoting.)
One more thing. Much of the research in this area measures back complaints rather than sleep, and that belongs with a clinician, not a sleep blog, so I’ve left it out. If your real question is about aches that keep coming back, take it to someone who can examine you.
“It takes about a month to get used to it”
If you’ve bought a mattress online, you’ve heard this one. The Sleep Foundation’s page on trial periods says it can take your body around a month to adapt, with break-in running two weeks to a month [3]. No study cited.
They may be right. In Jacobson’s study the scores were still climbing in week four. But that’s also exactly what novelty wearing off, or people simply settling in, would look like. I couldn’t find a controlled study that separates “your body is adapting” from “the thing is still new.” So: a decent rule of thumb, nothing more.
It does make you look twice at the fine print. The same page says trials average 90 to 120 nights, but 30-night trials turn up at budget retailers. If a month is how long it takes to settle in, a 30-night window leaves you about one honest week of judgment at the end, so a first impression deserves less trust than it’s going to feel like.
And the replacement rule? “Every seven to ten years” is everywhere, and I couldn’t trace it to a study. In Jacobson’s, the old beds averaged about nine and a half years, which tells you what people were sleeping on, not when a bed stops being fine.
Run your own trial
Now the part that isn’t research. If you suspect the mattress, collect some evidence before you buy anything.
Start by figuring out what you’re actually reacting to. At 2:15 a.m. it’s hard to tell the surface (sinking, an edge, a hard spot) from the heat from the person next to you rolling over. Those are three different problems, and only one of them is the mattress. If you’re waking hot, that’s a room-and-covers conversation. If your partner’s movement is doing it, that’s another one.
Your notes can help sort that out. If every bad night ends with the same sore spot on the same side, that’s a clue about the surface. If it’s always 4 a.m. and always hot, it isn’t. Add one line each morning about what woke you, and a week is enough to see which story you’re in.
Then borrow a different surface for a few nights, a guest bed, a friend’s couch if you’re brave, and write a few words each morning about how you slept. You’re not doing science. You’re checking whether the difference is big enough to notice without anyone telling you to look for it.
If you do buy, treat the trial like an experiment. The first week is unreliable, since new things feel like improvements, and a bad bed can fool you the same way in reverse. Keep the same morning notes for the first month and compare them with the old bed’s. If the notes didn’t change, that’s information too.
None of that has been tested as a method. It’s just what the holes in the studies suggest.
The loaf of bread
Next time you’re pressing a palm into the mattress at 2:15 a.m., ask what you’re hoping to find out. The mattress is easy to blame because it’s the one thing in the room you can point at. The studies suggest the real answer is smaller and less satisfying than the sales pages, and nobody can tell you in advance whether it’ll hold for you.
Diana Woods researches and writes about sleep, environment and everyday habit. She has no clinical training, and that shapes what this site does: instead of explaining what might be wrong with you, she reads the behavioural and environmental research and writes about the parts you can act on — the light in the hallway, the fifteen feet to the bathroom, what the morning does to the night that follows.
Every article here cites its sources, and she tries to include the parts researchers themselves flag as uncertain. Studies disagree, sample sizes are small, and correlation gets reported as cause more often than it should. Saying so is part of the job.
