Is it okay to sleep in separate bedrooms?

A 2023 survey by the American Academy of Sleep Medicine found more than a third of respondents said they occasionally or consistently sleep in another room to accommodate a bed partner[1]. That is not a fringe arrangement, and it is not confined to couples in trouble.

The number matters mostly because of what it does to the private version of this question. People asking whether it is acceptable almost always assume they are the exception, that everyone else is sleeping soundly side by side, and that admitting otherwise would be admitting something larger. The survey data says the opposite. A large minority of couples have already quietly made this adjustment.

What the surveys cannot tell you is how those couples are doing. Self-reported prevalence is not an outcome measure. Knowing that many people sleep apart tells you the arrangement is ordinary; it does not tell you that it works, and the two get conflated constantly in coverage of this topic.

Objective measures and subjective ones disagree here, which is the single most useful thing to know about this literature.

When sleep is measured with actigraphy or polysomnography, sharing a bed tends to look slightly worse: more movement, more brief arousals, more fragmentation. When the same people are asked how they slept, partnered sleepers frequently report sleeping better than they do alone. Both findings are real, and reconciling them is the actual research problem[2].

The leading explanation is that a partner in the bed provides something other than sleep efficiency. Reduced vigilance, a sense of safety, and the regulation that comes from a familiar presence all plausibly improve how rested someone feels without improving the polygraph trace. That would mean measuring only the objective side systematically undersells shared sleep.

None of this settles the individual case. If one partner has untreated apnoea, restless legs, a shift schedule, or is simply a very light sleeper, the objective cost is not marginal. It is chronic sleep loss, and chronic sleep loss is not a lifestyle preference.

The clearest evidence in this area is not about where people sleep. It is about what happens the next day when they have not slept.

Experimental and daily-diary work has found that poor sleep predicts worse conflict resolution between partners, more hostility during disagreements, and reduced accuracy in reading a partner's emotions[3]. The effect shows up on the days following bad sleep specifically, which is what makes it more than a correlation between grumpy people and grumpy relationships.

This reframes the trade-off. The question is not whether sleeping in the same bed is more loving. It is whether the version of you that shows up after six broken nights is the partner you want to be. For some couples the answer makes the decision straightforward.

It also explains why the arrangement can backfire. Separate rooms remove the sleep cost, but they also remove the default, unplanned contact that shared sleeping supplies for free: the conversation before sleep, the incidental touch, the shared waking. Couples who protect those deliberately generally do fine. Couples who assume they will happen anyway usually find they do not.

What decides this is not the sleeping. It is what the arrangement is being used for, and there are three quite different situations that look identical from outside.

The first is a scheduling problem. Different shifts, a snorer, a new baby, a partner with chronic pain. Here separate rooms are furniture logistics, and treating them as a relationship event adds a problem that did not exist.

The second is avoidance. One partner moves out of the bedroom during an unresolved conflict and the arrangement quietly becomes permanent, because it is easier than the conversation. The sleeping is not the issue and improving the sleeping will not touch it.

The third is a mismatch, where one partner experiences the arrangement as practical and the other experiences it as rejection. This is the version most likely to cause lasting damage, precisely because both people think they are talking about beds.

One further factor: reversibility. Couples who frame it as an experiment with a review date tend to report it going better than couples who frame it as a permanent verdict, partly because the first framing keeps it discussable and the second makes raising it again feel like an accusation.