My Old Dog Keeps Getting Up and Moving to a Different Spot at Night. Is the Bed Wrong, or Is It Pain?

It could be either, and one week of notes is how to find out which. Watch what happens after he moves. If he goes to one place and stays there for hours, he is choosing that place, and the reason is likely to be the bed or the room: too small, flattened under his hip, too warm, slippery beside it, or too far from you. If he lies down, shifts, gets up and does it again wherever he lies, the problem is travelling with him, and pain is the first thing to ask your vet about. Given a soft bed on the floor, nine old laboratory beagles in one small study used it for 83% of the night, so it is worth finding out why yours keeps leaving his. No study has shown that a new bed calms the night, including the one bed study that measured it. So start with the free changes, one a week, and buy only what the notes point to.

The bed is empty at 3am#

You bought the bed for him. It is in the corner where it has always been, and at three in the morning it is empty. He is on the kitchen tile, or across the bedroom doorway, or in the hall. Some nights he starts on the bed, gets up an hour later, lies on the rug, then tries the floor by the back door. In the morning you find him somewhere new and wonder whether you bought the wrong thing, or whether something is hurting him.

It is tempting to settle it one of two ways. "He just likes the floor," or "this bed is useless, we'll get a better one." Both answers skip a step. Where he goes when he leaves, and how long he stays there, is information, and it costs nothing to collect.

Does the problem stay behind, or does it come with him?#

This is the question to ask first, because the two answers send you in different directions. When he moves, does he settle?

If he goes to one place and stays, the thing he left is the problem: the bed, where the bed is, or what is around it. He has found something better and is telling you what it is. A dog who leaves foam for tile in summer and sleeps there until morning may well be shedding heat. Our elevated beds page covers why lying on a cool floor is one of the ways a dog cools down. A dog who leaves a bed in the spare room to sleep across your doorway is choosing company. A dog who leaves a bed he has to curl up on, to stretch out flat on the rug, has outgrown the bed.

If he lies down, shifts, gets up and does it again wherever he goes, the problem came with him. The floor is no better than the bed, and the rug is no better than the floor. That pattern is the one to take to your vet. It looks like a dog who cannot get comfortable anywhere, and the place he is uncomfortable may be his own joints.

None of these rows is a diagnosis. They are where to look first, based on the reasoning in each row rather than on any study, and a dog can have two at once. A dog who is hot and sore will still leave his bed for the tile, and the tile will not fix his hips.

dogorthopedic.com

What happens when old dogs are given an easy bed?#

Very little research looks at where dogs choose to lie, and almost none of it is in pet dogs with arthritis. What there is comes from laboratories, where researchers wanted to know what kind of bed dogs actually use.

In one study, nine elderly laboratory beagles in three groups were given beds filled with fleece blankets, placed either on the floor or raised. The beds on the floor were used for an average of 83% of the 10-hour dark period. Beds raised 30 cm (about 12 inches) were used for 21% of it. The authors wrote that the dogs "intensively used the resting places, especially at night, and rarely slept on the floor."^[2] Their earlier choice tests found a preference for soft bedding, and they recommended "a padded, soft bed which is easily accessible, especially for elderly dogs," with a variety of options, because the groups did not all prefer the same thing.

An earlier dissertation from the veterinary faculty at LMU Munich, on 21 laboratory beagles of unstated age, found a similar pattern: in its night observations, the dogs were on the beds provided for an average of 80.1% of the night, significantly more than on the ground. Fleece blanket was the favoured filling in every group, and a bed lined with a rubber mat was almost never used.^[3]

Be careful with these. They are small, they are beagles, they are laboratory dogs in kennels rather than pets in a house, and nobody checked their joints. But the point they make is the useful one here. In these kennels, old dogs given a soft bed on the floor stayed on it for most of the night. That is nine beagles, not a rule about every old dog. But it is enough to say that leaving the bed all night is not simply something old dogs do. If yours keeps leaving his, it is worth finding out whether the bed, the room or his body is the reason.

The raised beds are worth noticing too. A bed a dog has to step up onto was used for about a quarter as much of the night. Those dogs were not tested for arthritis, so this is not evidence about stiff dogs. It does fit a simple piece of reasoning: a dog who finds climbing hard has one more reason to pass a raised bed by.

Can a new bed change his nights?#

Nobody has shown that it can, and this is where the bed advertising runs ahead of the evidence.

The only bed study we know of that put activity monitors on dogs is the University of Pennsylvania pilot of the Big Barker bed: 40 dogs over 70 lb with osteoarthritis confirmed by a veterinary radiologist, all sleeping on the bed, with no comparison group and no blinding. The report's results page says that "50% of dogs had at least a 13% decrease in night-time activity and 25% of dogs had at least a 33% decrease," and calls this "suggesting improved nighttime restfulness."^[4]

The same PDF also has the statistician's appendix. It tests whether night-time activity at the end of the study was different from the week before the bed, in three ways: the third week's total against the baseline week, the average of all three weeks, and the third week's daily average. None of the three is anywhere near statistical significance: P = 0.67, 0.53 and 0.95. Daytime and 24-hour activity did not change significantly either. Both statements are true at once. In any group whose nights do not change on average, roughly half the dogs will move somewhat less and the rest somewhat more, just from ordinary variation. "Half the dogs moved at least 13% less" tells you nothing about the other half.

We had quoted the 13% line on four of our own pages, including this bed's product page, as evidence that the dogs moved less at night. That gave it more weight than the study's own appendix does, and we corrected all four today. The bed may still be a good bed, and the pilot's owner-scored results are a separate matter, covered on our bed guide. On the question of nights, the monitors found no change. As the next section explains, that does not prove the bed did nothing either.

There is one more useful number in that appendix. Before the beds arrived, the quietest dog's average night registered about 846 activity counts and the busiest about 21,500, roughly twenty-five times as much. Dogs differ enormously in how much they move at night. The only comparison that means anything is your dog against himself, which is why the notes below matter more than anything you can buy.

Why the monitors cannot settle it either way#

Here is the catch, and it cuts against the bed's critics as much as its makers. The same kind of collar monitor has missed a treatment that works. In the one study we found that asked owners about their dogs' nights, researchers first put collar monitors on 60 dogs with osteoarthritis for two untreated weeks and then two weeks on meloxicam, an anti-inflammatory painkiller. The monitors showed no significant change in night-time activity on the drug, and the authors concluded they did not appear useful for this. That first part had no placebo; each dog was compared with its own untreated weeks.^[5] In a smaller second part, 19 dogs were given the drug and a placebo in turn, and the owners, who did not know which was which, scored the dogs' sleep better on the drug. When another team from the same group reanalysed the monitor data from that crossover hour by hour, the treated dogs were less active between 11pm and 6am and more active in the morning, a difference the whole-night total had hidden.^[6]

So a whole-night activity total that does not move is weak evidence in both directions. It does not show that the Big Barker calmed those dogs' nights, and it does not show that it failed to. The honest summary is that nobody knows whether any bed changes an arthritic dog's nights, and the one study that tried could not have told us with the measure it used.

The owners' answers in that painkiller trial are worth one more look. The sleep questions that responded most to the painkiller were about twitching and dreaming, "followed by those asking about shifting position and vocalizing." The questions about moving around and pacing responded least.^[5] It was a small pilot of 19 dogs, and the questionnaire was new, so treat this as a hint, not a rule. It fits the pattern above: a dog who keeps shifting in place, wherever he lies, is the one most worth raising with your vet. A dog who walks once to a better spot and sleeps may be telling you about the spot, but that does not rule pain out.

“Because your dog can't tell you how much pain he or she is in.”

How do I tell which it is? The one-week notes#

Nobody has published a test for this, so here is a method. Keep these notes for seven nights before you change anything. A phone voice recorder on a shelf will give you most of the times. A camera tells you what he did when he was up.

At the end of the week, count two things. How many of his moves ended in a long sleep somewhere else? And how many were followed by more shifting? Mostly the first means the problem stayed behind. Mostly the second means it came with him.

dogorthopedic.com

Then change one free thing a week#

Change one thing at a time, and keep the notes going. If you change the bed, the room and his medication in the same week and his nights improve, you will not know which one did it. Do the free ones first.

The fist test. Press down hard with your knuckles at the exact spot where his hip lies, not the middle and not the edge. If you can feel the floor, so can he. Our bed replacement page explains this and the corner test in full.

The size check. Measure him asleep on his side, nose to the base of his tail, then add 8 to 12 inches (the working rule on our size chart, not a published standard), and compare that with the flat part of the bed, not the outside measurement with the rim. Our bed size page has the numbers by edition.

If he now stays on the bed, you have your answer and it cost nothing. If two or three weeks of changes make no difference and he still shifts and resettles everywhere, stop changing things. Take the notes to your vet, say "he cannot get comfortable anywhere, not just on his bed," and ask whether a pain trial is reasonable. The drug, dose and length of the trial are your vet's call. Keep the notes going through it, because that is how you will both know whether it worked.

When a new bed is worth it, and how to buy one so you can tell#

Buy a bed when the notes point to the bed: the fist test reaches the floor, he hangs off the end, or he curls on a bed he would rather stretch out on. Do not buy one because the nights are bad and you hope a bed will calm them. Nobody has shown that it will.

If your dog is under about 50 lb, you probably do not need a premium bed to fix any of those three problems. That is the cut-off on our Big Barker product page, and the reasoning is weight: a lighter dog is less likely to press a decent foam bed flat to the floor at the hip. A firm foam bed sized from the measurement above, or a foam topper cut to size inside a washable cover, does the job. Our bed guide lists cheaper options with their honest limits.

For a large dog whose notes say the bed has gone flat or is too small, this is the one we would buy.

What makes it fit this page is its return terms, not its foam. Bought direct from the maker, it comes with a 180-night trial. Read its conditions first: you cannot start a return in the first 30 days except for a size exchange, the request must be made within 180 days, you donate the bed to a registered shelter or rescue instead of shipping it back, and there is one trial return per household. Orders through other retailers, including Amazon, follow that retailer's return policy instead.^[7]

That trial is only worth something if you can tell, by night 30 or so, whether the bed has changed anything. That is what the week of notes before it arrives is for. Keep them before delivery and through the first month. If he now stays on the bed most of the night, you have your answer. If he leaves it exactly as he left the old one, you have a different answer, and the bed was never the problem.

What it is bad at. It is expensive: the Sleek edition, which has no headrest to step over, is $249.95 for the Large (48 × 30 × 7 in) and $299.95 for the Extra Large on the maker's site today.^[8] The card's Amazon button goes to the Headrest edition, which costs $40 more, and a bed bought through Amazon follows Amazon's return policy, not the 180-night trial; if the trial is why you are buying, use the card's direct button instead. No foam bed cools a dog the way a tile floor does, so it will not stop a dog who leaves his bed because he is hot. And its own study found no measurable change in night-time activity, as above. Buy it for the foam depth and the size, not for a calmer night, and read what its warranty actually says on our product page before you count on it. We earn nothing when you buy direct.

dogorthopedic.com

What to do this week#

Tonight, put your phone on a shelf with the voice recorder running and write down where he lies at lights-out. Keep the notes for seven nights. Do not buy anything yet. At the end of the week, count how many of his moves ended in a long sleep somewhere else, and how many ended in more shifting. If it is the first, move his bed to where he goes and see if he stays. If it is the second, book the vet and take the notes.

The delay this site exists to shorten often begins with a sign that gets explained away: he likes the floor, he is just restless, the bed is rubbish. Each of those may be true. But he is running the same test every night, trying one surface after another, and writing the result down takes a week. Our morning stiffness page covers what happens when he gets up, and how to tell if your dog is in pain covers the forms your vet will use.