I always get ads for fitness and health related products on my social media – it’s inevitable given what I look up when researching for articles! Lately I’ve been seeing one for a device that claims to be able to improve my posture, and reduce pain. It’s from Upright Technologies, and it’s called the Upright Go 2.

So today I’m going to review the Upright Go, its claims around posture and pain, and the evidence behind them.

To help, we’ve again recruited our good friend, physiotherapist Aaron Marshall. Aaron has contributed before when discussing the practices of physiotherapists and chiropractors, and is our go-to expert on musculoskeletal issues!

How does the Upright Go work?

You can stick the device to your upper back with an adhesive, or rest it there using a necklace (not included with your purchase), which is essentially a micro-USB cord the device plugs in to. It can detect movement in the same way your phone does. When you pass a threshold of movement in a certain direction (like dropping your shoulders forward when you slouch, causing the device to tilt), it will vibrate – again, just like your phone.

This vibration is a prompt for you to sit straighter. And this, according to ancient wisdom, results in less neck and back pain. Some even claim improved mood and confidence, among other benefits of standing up straight!

Sensibly, the advice is to build up the use of this device over time. This advice is “personalised” via answering a few questions on an accompanying app. Obviously you aren’t being assessed by a professional, so it’s unclear what formula they use to assess users. But that they are even trying to progress over time is good to see.

The app can be used to track your posture performance over time, and lets you know if you are improving.

How important is our posture?

It is not at all clear that posture is an important factor in pain, confidence, or even athletic performance. We’ll discuss some of this, especially around pain, later.

But it makes sense why we would think that. And for a long time, even the experts did. Aaron identifies that we have an ideal of what posture should look like, so anything that deviates from that “must be adding dangerous load to the spine, and causing pain”:

It is quite common from there to extrapolate to: altered mechanics lead to inefficient muscle activation, reduced strength and endurance, therefore less tolerance to these loads.

Our understanding of posture, and pain, have moved since these ideas. But not everyone will be up to date with this, so ideals around standing and sitting up straight are almost universally accepted by the public.

Can the Upright Go improve your posture?

I didn’t know what to think about these devices at first. I love the idea that they provide simple physical cues to improve our posture. And for the reasons discussed above, doing something to improve or maintain our posture seems like a good idea.

But though it sounds right intuitively, we need more information.

Of course, the website for this device certainly claims that it works. It does so through the usual mixed bag of infographics, testimonials, and some select pieces of research.

The research we’ll review in detail later. But I’m not often convinced by testimonials.

Wait, why don’t you like testimonials?

These guys use a lot of testimonials on their website! There are thousands of them.

Why aren’t I convinced by these? Because I don’t know if what the users report is a real effect, or just because they expect to feel different. Any intervention works better than no intervention at all, due to the placebo effect.

And the nature of chronic neck and back pain is repetitive. Our pain will often come and go over time. When it hurts, we might look for something to make it better. That might be stretching. It might be light exercise, chiropractic, or watching cat videos online.

When the pain eventually goes away, we fall victim to an attribution bias, and assign credit to whatever fits our values: “those cat videos really worked wonders for my neck pain!”.

Depending on the nature of the pain, sometimes the intervention we choose (exercise, for example) does help. But it’s hard to know, because we don’t have anything to compare it with. We think we would have been in pain for longer without the intervention, but we don’t know.

So a testimonial doesn’t really help. Thousands of testimonials don’t help either, because they all have the same problem.

Proving that a new product works more than a placebo does can be difficult, and time-consuming. This is why we do research.

Let’s review the research. Can the Upright Go reduce pain?

There are three pieces of research listed on the website as proof this device works. I’ll do a deep dive into all three.

The first study to review lists Professors Richard Harvey and Erik Peper as the first two authors. Neither of these two have training in biomechanics or physiology. Instead, both have doctorates in psychology. However Peper has a long history of research in this area, looking at the role of posture on thinking, and the interaction between health and technology.

How was this study conducted?

There were 59 volunteers for this research, of which 13 self-selected to join the intervention group that would be wearing an Upright Go device. Another 13 were selected from this group to be a control group.

An immediate problem is these groups are not the same, even if the demographics of the groups are identical. The people volunteering to join an intervention group want to improve. They will expect and welcome change. The control group were happy to do nothing.

This brings us to a second flaw in the design… using of a control group, instead of a comparison group (which could receive, for example, education about good posture). As discussed above, any intervention is often better than nothing, even if just because we expect it to work.

So here we have a motivated group of people, exposed to an intervention (the Upright Go). And we have a less motivated group that did nothing. I would be shocked if the intervention group did not improve!

Perceived measures of health, energy, well-being, social functioning, and pain were gathered before and after 4 weeks of regular use of the device. Participants also reported on their sleep quality, and how strong they felt.

That’s right, how strong they felt. They didn’t assess the actual strength of participants!

Erik Peper has a history of poor research.

If I had proposed assessing strength like this to any of my research collaborators, I would have been laughed out of the room. But it seems Peper has never tried to accurately assess strength. In some 2016 research he used a manual muscle test often used by quacks like “applied kinesiologists”, or “chiropractors” to assess the strength of participants in another study. Again, instead of objective strength testing!

He also researches another mode of quackery known as “therapeutic touch”, which proposes that someone can be healed through manipulating their energy fields with your hands. Energy fields that cannot be observed or measured, by the way, but hey, let’s just assume they exist anyway!

I have some serious concerns that an experienced professor would produce research of this quality.

But for the record, it should be noted that I’m not calling Erik Peper a quack researcher. A quack researcher would do a lot of the things that Erik Peper does, but you can form your own opinion.

But it gets even worse!

If you thought this study couldn’t get any worse, there is another problem. Participants also received instruction on the importance of posture, and participated in the subjective, manual muscle testing mentioned above to show them how strong they were with better posture.

The paper isn’t completely clear about this, but it seems like this instruction was only given to the intervention group. So not only was the experimental group self-selected (so already more motivated than the controls), the researchers interfered by making them even more motivated to improve their posture. Now we can’t tell how much of any improvement came from the device, and how much came from the instruction.

The intervention group improved, but not in all measures. They reported less fatigue, more confidence, and less stress. But pain didn’t improve significantly – one of the key claims of this device.

So the first of three key pieces of research highlighted by the Upright Go website didn’t support one of their key claims. Not a great start, is it? Let’s review the others.

The second study used to support the Upright Go

The next piece of research we will review is a Master’s thesis completed in 2018, that has not had any findings published in an academic journal. One of the supervisors for this work was Erik Peper, a senior author on the paper discussed above.

This study aimed to assess change in the kyphotic curve of the spine due to wearing the Upright Go – this is the convex curve that we all have in the mid portion of our spine. A large kyphotic curve can result in the forward shoulder, slumped posture we often think is “poor”. They also proposed that confidence and self-esteem would be improved from using this device.

The problem here is that this study doesn’t provide any evidence that pain will be reduced by using this device. It just assumes that pain will reduce if appropriate changes are seen in the spine of participants who use this device.

So this is another bad piece of research?

It sure is! When writing a thesis the student first needs to prove that the kyphosis they are looking to achieve with this device is associated with less pain, in a detailed review of the research to date.

They didn’t do this. The “detailed review” is only 6 double-spaced pages long. There was only one study cited as evidence that slouching is a cause of pain. And this study was left off the reference list, so I can’t read it myself.

The review of the research was sloppy in other ways. This statement, for example, doesn’t give me much confidence:

In female college students, carrying a backpack increased their angular pelvic tilt and range of motion was decreased, suggesting that backpack carrying could cause permanent posture deviations (Smith et al., 2006).

How it can simply be assumed that the change in posture when carrying a backpack can lead to permanent change is not clear. The study cited here was also missing from the reference list.

These sorts of wild claims, sloppy referencing, and weak assessment of the research is understandable in a student learning the research process. But the supervision committee should pick these things up, and not let bad work get submitted. If they don’t pick them up, the examiners definitely should.

The review process has let us down in this case!

Was there any benefit to using this device to change posture?

There was no improvement in either the measures of self-esteem, or confidence, when using the posture training device.

There was minor change in the seated posture of the experimental group. Which you would expect when someone is being assessed on their posture, and after weeks of posture training with this device! But still no evidence there was any effect on pain! Again, this is a key claim of the device!

The thesis finishes by claiming that future spinal problems can be reduced by using these posture trainers… but this is speculation. They didn’t prove this.

Hopefully this final piece of research is better!

It’s not, sorry. The last piece of research we will review, which is claimed to support the use of this product, is underwhelming, to say the least. It’s a poster – that’s right, a poster – presented at an academic conference.

To be fair, academic posters can be impressive, summarizing a new piece of research for interested conference attendees.

But these posters are usually work that has not yet been published, so are not reviewed by other scientists. And details are usually minimal.

In this poster it appears as if there was no comparison group at all, not even a control group. And while the participants changed their posture when using the device, a link between this and pain was not established.

Even more important, these participants were sufferers of Parkinson’s disease. They weren’t the typical neck and back pain sufferers that this product is marketed to!

In research we can only confidently apply our findings to the group of people we have tested. If we want to know if a product has a broader effect, we need to test it on more people.

What do we think of the research we’ve reviewed?

Honestly, not much. The company has used three different pieces of research to support the use of their product. Two of them have not been published, and the other was so poor I’m surprised it was published.

None of them demonstrate that pain is reduced, a central claim of the product. One showed no decrease in pain, and the other two did not assess pain.

Presumably, this is the best they can offer! If there was research more supportive of their product, I have no doubt they would use it.

Misrepresenting research, cherry-picking only the research which supports them, or relying on poor quality research, are all hallmarks of quackery. I suspect that this product has come to market before the manufacturers had any real evidence that it was effective.

But they wouldn’t be able to find good research, as Aaron identifies:

there is limited good quality research into the principles of ergonomics, including posture and pain. A couple of comments from highly respected clinicians really stand out. Pete O’Sullivan confirmed the evidence for this field of inquiry was “…built like a house of cards.” And Associate Professor James McAuley less subtly stated “I think it is a load of rubbish.”

So what does an impartial reading of the research tell us?

Essentially, there’s no clear relationship between posture and pain. Some of us are in pain. Some of us have postures that we traditionally consider “bad”. But these often are not the same people.

Some academics (such as Erik Peper) assume that poor posture causes pain, then conduct their research based on this assumption. That’s one hell of an assumption to make!

I’m not going to review the research exhaustively here, because Paul Ingraham at Pain Science has already done a brilliant job of it. But put simply, there is no consensus in the research.

• Posture is highly variable. In fact, it even varies for the same person between measurement.
• The curve in our lower back, the lordosis, is also highly variable, as is the amount of movement in our lower back. Some research shows less movement in those with lower back pain than those without, but this could be a guarding response, due to pain. It’s not been established as a clear cause and is contradicted in other research.
• Australian research showed no association between neck posture and pain in adolescents. Other research confirmed this, but did find an association in older adults.
• Thoracic posture is not related to shoulder pain.
• There is no clear association between cervical posture, and temporomandibular joint pain, shown in multiple reviews.

The interaction between pain and posture sounds complicated!

It is complicated. One of the problems with taking a biomedical approach to pain, as Aaron has discussed with us before, is that pain is more complex than just anatomy:

Pain is multifactorial. Stimulus from many different facets of life, like stressors (emotional, relationship, financial), family dynamics, health beliefs, spiritual beliefs, physical damage or potential damage, could all contribute. Couple this with a lack of reliability and validity in assessing posture, there can sometimes be an association between posture and pain, but not a consistent one. And certainly not a cause-and-effect relationship.

When reviewing research you look for a consensus starting to form as you immerse yourself in the research. If you can’t see a consensus, there isn’t a relationship. So we can’t assume there is.

Then even after all that, if there is a relationship between posture and pain, it needs to be shown that this device can change your posture in a lasting way. Then it needs to be shown that this leads to a reduction in pain.

This would take years of research. But when you have a product to sell now, why bother with high-quality, long-term research? Instead, just slap on your website some shoddy research, from shoddy researchers, with only the slightest relevance to the claims you make.

But this will be convincing enough to the general reader, with no knowledge in the area.

What’s our verdict on the Upright Go?

It overpromises, and underdelivers.

Wearable devices like this are a great idea. Research shows they can reasonably accurately measure changes in your posture as they occur. And the feedback provided can change your standing or seated posture, at least in the moment.

But this doesn’t mean that you can sustain these changes, or that you feel less pain as a result.

Aaron also points out that the benefit of this device could be recreated with a cheap piece of strapping tape. After all, the proposed benefit is from physical feedback.

While these devices may prompt people to achieve a more upright posture (which may be helpful in some circumstances) I can’t, for the life of me, think of a time where a costly device would be my “go-to”, over a small piece from a roll of rigid strapping tape.

So we think if you want to try them, go for it. And if you can afford it, great. But they probably don’t live up to the claims.