Mattress firmness

Medium-firm is the best-supported choice, especially if you have low back pain. The evidence is real but low quality, so your own comfort over a few weeks matters most.

• Evidence: Weak
• Medium effort
• High cost
• Works: about 4 weeks

Why it matters

Your mattress holds you in one position for hours. Too soft and your hips sink, bending your spine. Too hard and your shoulders and hips carry the pressure. Reviews of mattress studies suggest a surface somewhere in the middle gives better comfort, sleep quality and spinal alignment [1][2].

The firm-is-best idea is old and was never well tested. In a trial of 313 adults with chronic low back pain, people given a medium-firm mattress had less pain in bed and less disability after 90 days than people given a firm one [4]. The “firm” mattress in that trial was close to the hard end of the scale used to rate firmness, so the result argues against very hard beds, not against any firmness.

How to do it

1. Start at medium-firm. That’s the most defensible general choice for adults with nonspecific low back pain or morning stiffness [3].
2. Adjust for your body and position. The 2026 review suggests also considering your body size, age, preferred sleep position and how warm you sleep [3]. As a rule of thumb, side sleepers usually want enough give at the shoulder and hip, and heavier people usually need more support. See sleep positions.
3. Test it properly. Lying on a mattress for a few minutes in a shop tells you little. Look for a home trial period that lasts at least a month, with a clear and cheap return process. In the new-bed studies, people tracked their sleep for 4 weeks and kept reporting improvements through that time, so judge after weeks, not nights [5][6].
4. Don’t trust labels across brands. One seller’s “medium” can be another’s “firm”. Studies themselves rated firmness subjectively, which is part of why the evidence is messy [1]. Compare mattresses by lying on them in your usual sleep position for several minutes each.
5. Couples with different needs: look for a mattress with different firmness on each side, or two single mattresses on one frame.
6. Replace when it fails, not by the calendar. Signs: a visible dip, springs you can feel, waking stiff and loosening up soon after you get up, or sleeping better in hotel or guest beds. In two small studies, people whose own beds averaged 9.5 years old reported better sleep and less back pain on new medium-firm beds [5][6]. There’s no good evidence for a set replacement age.

A topper can make a firm mattress softer for much less money. It can’t fix a mattress that sags.

Common mistakes

• Buying the firmest mattress “for your back”. The best trial found the opposite in people with back pain [4].
• Expecting a new mattress to cure back pain. It may help comfort. If your back pain doesn’t improve after a few weeks, is worse at night or when resting, or stops you doing daily things, see a GP. Pain, tingling or numbness in both legs, or changes in bladder or bowel control, need emergency care [7]. More on when to see a doctor.
• Judging on the first night. Any new surface feels odd. Give it a couple of weeks.
• Paying for features with no evidence. Self-adjusting air mattresses and temperature-controlled mattresses showed promise in some studies, but the reviews call the evidence insufficient or say larger trials are needed [1][3].

Effect on your sleep score

We found no study measuring the effect of mattress firmness on wearable sleep scores in healthy adults.

• Sleep quality and efficiency: the new-bed studies reported better self-rated sleep quality and efficiency [5][6]. These were self-ratings, with no control group, so some of the gain may be the novelty of a new bed.
• Wake-ups: if pain or stiffness wakes you, a better-matched mattress could reduce wake-ups. That’s plausible, not measured.

How confident: low. Don’t expect a visible change in deep sleep, REM or HRV.

Evidence

Rated Weak. The direction is consistent: medium-firm beats very firm or very soft in three systematic reviews of 24, 39 and 47 studies [1][2][3], and in one well-run randomized trial [4]. But:

• The reviews didn’t pool results in a meta-analysis, because the studies used different mattresses, measures and follow-up times [3]. The 2026 review grades its own evidence as level III [3].
• “Medium-firm” was often defined by how it felt, not by a standard measure [1].
• The only large randomized trial was in people with chronic low back pain, not healthy sleepers [4].
• The new-bed studies compared people’s old beds with new ones in the same people, so they can’t separate firmness from newness [5][6].

General information for healthy adults, not medical advice. If you have a sleep problem that lasts, read when to see a doctor. Sources checked 4 October 2026. Spot a mistake? Tell us.