① The claim that bending your head puts kilos on your neck
One of the things that made people start watching the angle of their neck was a number: bend 15 degrees and your neck carries 12 kilos, or something along those lines. I went looking for where it came from. Nobody put a scale under anyone's neck — it's a computational model published in 2014.
The 2025 paper I read for this piece refers to that study directly. It calls it a computational model that proposed the premise that the more you flex your neck, the greater the load, and notes that it rests exclusively on biomechanics without accounting for the fact that pain is modulated by many factors.
That isn't to say the arithmetic is wrong. But the load a model calculates and who actually ends up in pain a year later are different questions. So I went looking for research on the second one.
② They followed 396 people without neck pain for a year
A Brazilian team recruited adults aged 18 to 65 who had no neck pain, and measured the cervical flexion angle they adopted while texting on their phones. Not a questionnaire asking "do you look down a lot" — an actual inclinometer reading.
▸ 457 recruited at baseline, 396 (87%) completed the one-year follow-up
▸ Measured angles averaged 34 degrees standing, 36 degrees sitting
▸ At one year, 40 people reported neck pain — about one in ten
Then they checked whether that angle predicted who hurt a year later. Standing and sitting were analysed separately, and the results were nearly identical: odds ratio 1.01, confidence interval 0.98 to 1.04. In other words, no basis for saying that one more degree of flexion changed anyone's odds. The same held when they asked about how often the pain occurred.
③ What did move with the pain was sleep and activity
The same analysis looked at other factors, and two of them tracked with neck pain.
▸ Poor sleep quality — odds ratio 1.76 (confidence interval 1.17 to 2.63)
▸ Insufficient physical activity — odds ratio 2.41 (confidence interval 1.03 to 5.65)
These need a particular way of reading. The starting point is that one in ten people reported neck pain over the year, and an odds ratio is not the same thing as "this many times more likely to hurt."
The activity figure especially: its confidence interval runs from 1.03 all the way to 5.65. It clears 1 by a hair, so the direction shows but the size stays blurry.
One more thing worth putting plainly. The group with the least activity — the "sedentary" category — was not statistically significant (odds ratio 2.39, confidence interval 0.83 to 6.86). The authors themselves went no further than saying it is suggested. For those of us sitting all day this is the most tempting line in the paper, and it is also where the evidence is thinnest.
④ And yet there is research saying heavy phone use hurts your neck
Stopping here would show you only one side, so I went looking for the other. A meta-analysis published the same year pooled 7 studies and 10,715 participants and reports an odds ratio of 2.34 (confidence interval 1.44 to 3.82) for neck pain among people who overuse smartphones.
On the numbers it looks like a head-on collision with the study above. But reading both in full, they measured different things.
▸ The meta-analysis measured usage time and usage habits. The study above measured the angle of the neck
▸ The studies pooled in that meta-analysis were, as its own authors state in the limitations, all cross-sectional. They asked "does your neck hurt right now" and "how much do you use your phone" at the same moment, which cannot separate hurting because you used it from using it less because you hurt
▸ The variation between studies is enormous (I² of 94%). What counted as overuse differed from paper to paper, and one of them used a threshold of more than 20 text messages a day
▸ On quality assessment, 5 of the 7 studies scored only one point in the outcome domain, because they relied on self-report and the follow-up was not long enough
Most tellingly, the authors of that meta-analysis close by saying that longitudinal studies are needed to settle this. That longitudinal study is the one in section two — and what separated people there was not the angle, but sleep and activity.
⑤ What this study did not test
The things I kept flagging as I read.
Sleep quality was captured with a single question. Not one of the established sleep questionnaires, but "did you have trouble sleeping in the last month?" with four response options. That is enough to see a direction, but not enough to tell you which aspect of sleep is the problem.
The authors also qualified how the angle was measured. It remains unknown, they write, whether the neck posture participants adopted during the study task differs meaningfully from their habitual posture when texting. What was measured is the angle in that moment, not the angle across a day.
Participants averaged 27 years old and 70% were women. The stated range was 18 to 65, but in practice the sample skewed young. The follow-up period overlapped with 2020, and the authors note people may have been more sedentary than usual.
And this is an observational study. Nobody was randomly assigned to bend their neck, so the most that can be said is that things were associated.
⑥ So what do you do with this
What this research offers is a direction, not an angle to hit.
▸ If your neck doesn't hurt right now — there is nothing in this data to justify spending time measuring angles or memorising corrective postures. Looking at sleep and activity is where this study points
▸ If you've been holding back because you worry about looking down — the angle itself did not predict pain a year later. That said, this study does not go as far as "so bend as much as you like." What was measured is the angle in a moment, not how many hours you stayed in it
▸ If your neck already hurts — this study followed people who started without pain, so it says nothing about treatment. If the pain persists or numbness travels down your arm, see a clinician
A line about how we read this while building BaroSit. What came out of it isn't a target number of degrees; it's an axis — how long you stayed put, and how much you moved. That's the side our nudges are on.
Change the question and it gets easier to answer. Not "what's my angle," but "am I sleeping well lately, and how much am I moving?"
We've written separately about the link between posture and pain, and about sleep and activity each in their own right.
→
barosit.com/community/p/b7d2e9a4-1c03-4e58-9f21-0a6b3c8d5e11
→
barosit.com/community/p/99e4a6d7-ba38-485e-a166-7e942d9ae3c6
→
barosit.com/community/p/d5f1a8c3-9e46-4b28-8c73-1a6f3d9e5b20
Sources
▸ Correia IMT, Ferreira AS, Gomes JFM, Reis FJJ, Nogueira LAC, Meziat-Filho N. Cervical flexion posture during smartphone use was not a risk factor for neck pain, but low sleep quality and insufficient levels of physical activity were. A longitudinal investigation. Braz J Phys Ther. 2025;29(6):101258. Twelve-month prospective observational study. 457 adults aged 18–65 without neck pain recruited; 396 (87%) followed up. Cervical flexion measured at baseline with a cervical range of motion device (34°±12 standing, 36°±14 sitting). At one year, 40 (10%) reported neck pain. Text neck odds ratio 1.01 (0.97–1.04) standing and 1.01 (0.98–1.04) sitting; sleep quality 1.77 (1.18–2.64) and 1.76 (1.17–2.63); insufficient activity 2.46 (1.05–5.75) and 2.41 (1.03–5.65); sedentary 2.39 (0.83–6.86), not significant. No competing interests; funded by Brazilian public research agencies (FAPERJ, CAPES).
▸ Chen YJ, Hu CY, Wu WT, et al. Association of smartphone overuse and neck pain: a systematic review and meta-analysis. Postgrad Med J. 2025;101(1197):620–626. 7 studies, 10,715 participants; adjusted odds ratio 2.34 (1.44–3.82), I² = 94%. The authors state in their limitations that all included studies were cross-sectional, that overall quality was moderate, and that 5 of the 7 scored only one point in the outcome domain. They identify the varying definition of overuse as a source of heterogeneity and call for longitudinal studies in their conclusion. No competing interests.
▸ The 2014 computational model referred to in the text was not obtained in full; it is discussed here only as characterised by the 2025 paper above.
This article is not medical advice. If neck pain lasts more than two weeks, if numbness or weakness travels down your arm, or if it began after a fall or accident, see a clinician. This study followed people who started without pain, so it does not speak to treating pain you already have.