Does sleeping with a pet actually hurt your sleep quality and HRV?

What does the actual sleep-lab data say?

The most-cited objective study here is a 2017 Mayo Clinic Proceedings paper that tracked 40 healthy adults without diagnosed sleep disorders and their dogs over 7 nights each, from August to December 2015. Both human and dog wore validated accelerometers rather than relying on self-report. Average human sleep efficiency across the whole sample was 81% ± 7% — already a touch below the roughly 85% often cited as a healthy target — and dogs averaged 85% ± 15%.

The key finding wasn't "dog in the room, yes or no." It was where the dog slept. Sleep efficiency was significantly lower when the dog slept on the bed (about 80%) compared with in the room but off the bed (about 83%), a statistically significant difference (p = .003). The authors' actual conclusion ran against the popular assumption: a dog's presence in the bedroom, by itself, did not look disruptive to human sleep. The disruption showed up specifically when the dog shared the mattress.

A pet is one of several bedroom-level disruptors worth checking against your own data alongside things like traffic noise, which can measurably disturb sleep even when it never fully wakes you. A smaller, independent actigraphy study of 12 dog-owning women added a texture to this: dogs' nighttime movements measurably predicted the women's own movements more than the reverse, yet the women were largely unaware of it — they rarely blamed a restless night on the dog in their own sleep diaries, even when the accelerometer data showed the dog moving around them. That mismatch between what people notice and what a tracker records is a theme that runs through this entire topic, and it's the same mismatch covered in why your recovery score can disagree with how you actually feel — the instrument and the subjective report are measuring different things.

That last row matters for how much confidence to put in any single number here: two different self-report surveys, asking similar questions, landed on opposite conclusions about dogs. The honest summary is that the objective, device-measured evidence (the Mayo study, the actigraphy studies) consistently points to a small cost specifically tied to on-the-bed sleeping, while the self-report survey evidence is genuinely mixed and sometimes contradictory.

Does a pet in the bedroom — not on the bed — still matter?

Based on the strongest available evidence, no, not in any way these studies could detect. The Mayo study's own framing is worth repeating because it cuts against the instinct to treat "pet in the bedroom" as inherently disruptive: a dog present in the room but off the bed produced sleep efficiency close to what you'd expect without a dog at all. The disruption was specifically tied to a 60+ pound (or any size) animal sharing the same sleep surface — shifting positions, taking up space, radiating heat, or occasionally needing to be let out.

This is a genuinely useful distinction if you're trying to decide between "pet sleeps on the bed," "pet sleeps on the floor of the bedroom," and "pet sleeps in another room," because the data suggests the middle option captures most of the companionship with little of the measured cost.

Does a cat behave differently from a dog?

The pattern in self-reported data says yes, though this rests on weaker evidence than the dog findings. The 2024 Trinity College survey of roughly 1,500 US adults — about half of whom reported co-sleeping with a pet — found dog co-sleepers reported worse perceived sleep quality and greater insomnia severity compared with people who didn't co-sleep with a pet. Cat co-sleepers showed no meaningful difference from non-co-sleepers, and in some of the study's exploratory analyses, cat co-sleeping was associated with slightly better self-reported sleep efficiency.

Two caveats are worth being upfront about. First, this is a self-report survey, not an accelerometer or polysomnography study — it captures what people believe about their own sleep, not what their sleep actually looked like that night. Second, the authors describe the species-specific breakdown as exploratory, not confirmed, and it hasn't been independently replicated with objective measurement the way the Mayo dog study has. A plausible but untested explanation is that cats are typically smaller, move less overnight, and don't need middle-of-the-night bathroom breaks.

Why do most people feel fine about it even when the data looks worse?

This is arguably the most interesting number in the Trinity survey: 93% of pet co-sleepers believed their pet had either a neutral or positive overall effect on their sleep — reported in the same dataset where dog co-sleepers, as a group, showed worse perceived sleep metrics than non-co-sleepers. That's not necessarily a contradiction. A pet can genuinely reduce loneliness, provide a sense of security, and make falling asleep feel easier — real psychological benefits that a sleep tracker or a sleep-efficiency percentage was never built to capture — while the same body is getting measurably more fragmented, shorter sleep on the instrument that does measure it.

This is the pet-specific version of a pattern this site has covered before in the context of training: your subjective sense of a good or bad night and an objective measurement of the same night can both be true and still disagree, because subjective and objective recovery signals pick up different things. Neither one is "wrong" by itself; they're answering different questions.

What about kids?

The evidence in children and adolescents looks less consistent than the adult dog data, and it's worth knowing both results exist rather than picking whichever one confirms what you already believe. A pediatric polysomnography study — a stronger method than actigraphy alone, since it directly measures brain-wave sleep stages for one night, backed by two weeks of actigraphy and diaries — found sleep profiles largely similar whether or not children shared a bed with a pet, with one exception: frequent co-sleepers had a longer PSG-measured sleep-onset latency than occasional co-sleepers. A separate, more recent actigraphy study of 175 families reached a different conclusion for the pet-on-bed subset specifically: shorter total sleep, lower efficiency, and longer sleep-onset latency versus no pet in the bedroom — though the dog-owning sample was only 43 families, and the authors flagged the result as needing replication. Awakenings and self-reported complaints didn't differ between groups in either study. The practical read: small samples, imperfect agreement, no reason for alarm, but no basis for the strong claim that pet co-sleeping is definitively fine in children either.

Is there any human HRV research on this specifically?

No — this is worth saying plainly rather than implying otherwise. A search of the available sleep-science literature on pet co-sleeping turns up sleep efficiency, sleep duration, sleep-onset latency, and self-reported sleep quality as the outcomes researchers have measured. None of the available human studies measured heart rate variability or other autonomic nervous system markers during pet co-sleeping. (There is HRV research on dogs' own sleep physiology, which is a different question entirely and not something that tells you anything about the human sleeping next to them.) If an article or forum post claims a specific HRV number tied to pet co-sleeping, treat it skeptically — it isn't backed by a published study as of this writing.

That gap is exactly where tracking your own data earns its keep instead of waiting on a large enough study to get published.

A 2-week self-test protocol

Because the published evidence only measured sleep efficiency and duration, and your bedroom and pet don't match any study's average participant, the most useful thing you can do is run a short, structured comparison on yourself using whatever sleep and HRV tracking you already have.

1. Pick three conditions and commit to at least 4 nights each, ideally over 2 weeks: pet on the bed, pet in the bedroom but off the bed (a dog bed or crate works), and pet in a separate room entirely. Keep everything else as constant as you reasonably can — same bedtime window, similar alcohol/caffeine intake, similar stress level.
2. Track three numbers each morning: your overnight HRV, resting heart rate, and sleep efficiency (time asleep ÷ time in bed) if your device reports it. Don't rely on memory of how the night "felt" — that's exactly the signal the Trinity survey shows can mislead.
3. Look at trend, not single nights. One bad night proves nothing; a consistent 3-4 night pattern in one condition is a real signal. If your device shows a rolling HRV trend, compare the average across each condition block rather than night-to-night noise.
4. Separate "the pet moved the bed" nights from "normal" nights if you can — a quick note in a sleep diary ("let the dog out at 3am," "cat jumped on/off twice") turns an ambiguous bad night into an explained one, rather than a mystery data point.
5. Decide based on your own numbers, not the published averages. The Mayo study's 3-percentage-point difference is a population average across 40 people; your own pet, your own bed size, and your own sleep anatomy could easily show a bigger effect, a smaller one, or none at all.

If the pet-on-bed nights show a consistent efficiency or HRV cost in your own data and you're comfortable with a crate, bed, or closed door for your pet, the evidence suggests the room-but-off-bed arrangement keeps most of the companionship with little of the measured downside. If your own numbers show no real difference, that's a valid and common outcome too — the Mayo study's finding was a group average with real individual variation underneath it, not a universal rule.

When this isn't just a sleep-optimization question

None of this evidence suggests pet co-sleeping is a health risk that needs medical attention — at worst, the documented effect is a modest reduction in measured sleep efficiency, not a safety concern. The one situation worth treating differently is a known pet allergy or asthma that's being aggravated by close nighttime contact with dander; that's a conversation for a doctor or allergist armed with your symptom pattern, not something a sleep tracker's efficiency percentage will diagnose. For everyone else, this is squarely a personal-preference-versus-data trade-off, and the self-test above will tell you more about your specific situation than any published average can.