Is sitting the new smoking? Why the hours are not the signal, but the metabolic stillness

Vallance and an international team from Athabasca University, Melbourne and Brisbane placed the most current estimates of the time next to each other. For a lot of sitting, the relevant meta-analysis found a risk ratio of 1.22 for all-cause mortality, with a confidence interval from 1.09 to 1.41. For current smokers compared with people who had never smoked, the relative risk was 2.80 for men and 2.76 for women. With more than 40 cigarettes a day it was 4.08 and 4.41.

It becomes even clearer in absolute numbers. The authors attributed more than 2000 additional deaths per 100,000 people per year to the heaviest smokers. For the comparison of the highest with the lowest sitting time it was 190. That is roughly one tenth. For you that means: sitting remains a risk factor to take seriously, it is simply not smoking.

Biswas and a team from Toronto evaluated 47 papers and adjusted for physical activity in the process. For long sitting time an association with all-cause mortality remained, with a risk ratio of 1.240. For newly developing type 2 diabetes the value was 1.910. What stood out was this: the associations were consistently stronger in people who moved little than in very active people.

Patterson and a team from the University of Cambridge and Imperial College London pooled 34 prospective studies with 1,331,468 participants and looked for the shape of the dose response curve. The result was not a straight line. Below roughly 6 to 8 hours of sitting a day, mortality risk barely rose statistically. Above that it increased by around 4 per cent per additional hour. For television time the threshold was at 3 to 4 hours and the rise was steeper.

For you that means: it is not about every single minute of sitting. It is about whether your day overshoots that range.

Ekelund from the Norwegian School of Sport Sciences in Oslo and an international consortium harmonised the raw data of 16 cohort studies. 1,005,791 people, 2 to 18 years of follow-up, 84,609 deaths. In the most active quartile of participants, which corresponds to roughly 60 to 75 minutes of moderate activity a day, sitting time was no longer associated with increased mortality. People who sat for more than 8 hours and were at the same time very active were at a risk ratio of 1.04.

People who by contrast sat little, under 4 hours, but were in the lowest activity quartile, had a risk ratio of 1.27. Put differently: sitting little and barely moving was less favourable in these data than sitting a lot and moving a lot. For television time, however, a residual risk remained even with high activity.

Stamatakis and a team from the University of Sydney followed 149,077 adults for a median of almost nine years. Here too the pattern showed up: in people who reported no moderate to vigorous activity at all, mortality risk rose almost stepwise with sitting time, from under 4 hours to over 8 hours with a risk ratio of 1.52. In those who met the activity recommendations, the association was weak and inconsistent.

Bey and Hamilton specifically prevented rats from their normal, everyday moving about, without training them and without starving them. Within a short time roughly 90 to 95 per cent of the heparin-releasable LPL activity in the muscle was lost. In parallel, less labelled fat was taken up from the blood into the muscle, and HDL cholesterol fell.

One detail was decisive: the amount of the building instructions, meaning the LPL messenger RNA, stayed unchanged. So the body had not cancelled the order for the enzyme. It had taken it out of service. Four hours after light treadmill walking was resumed, activity was around eight times higher.

In a review in Diabetes, Hamilton and colleagues summarised what follows from this. Experimentally reducing normal standing and walking time had, in their investigations, a larger effect on LPL regulation than adding intense training on top. And the cellular processes during inactivity differed qualitatively from those during training. Their conclusion: too little everyday movement and too little exercise could be two different problems that do not cancel each other out.

In an earlier review, Hamilton, Areiqat, Hamilton and Bey described an additional subtlety. Not all muscle fibres responded in the same way. For high LPL activity in the slow, posture stabilising muscles, everyday walking and standing were what mattered most. The fast fibres, by contrast, needed more intense contractions. For you that means: your postural muscles are spoken to by light continuous movement, not by interval training in the evening.

Krogh-Madsen and a team from the University of Copenhagen asked ten healthy, non athletic young men to deliberately lower their step count for two weeks. From an average of 10,501 to 1,344 steps a day. They did not diet, they did not fall ill, they did not stay in bed. They simply walked considerably less.

Afterwards glucose uptake in the hyperinsulinaemic clamp test was 17 per cent lower. The cause lay in the periphery, meaning in the musculature, not in the liver. Maximal oxygen uptake fell by 7 per cent. And the fat free mass of the legs had decreased, that of the arms and trunk had not.

For you that means: the body saves very quickly on what it no longer needs. Two weeks were enough for a measurable signal.

Morishima and a team from the University of Missouri had eleven young, healthy people sit for three hours. One leg was moved regularly, one minute of fidgeting, four minutes of rest. The other leg stayed still. In the resting leg, shear rate in the popliteal artery fell from 71.7 to 20.2 per second, and flow mediated dilation fell from 4.5 to 1.6 per cent.

In the moving leg, shear rate rose briefly to over 220 per second during fidgeting. And after the three hours vascular function was not worse but higher than at the start, at 6.6 instead of 3.7 per cent.

For you that means: it was literally the same blood, the same blood sugar, the same stress level. The only difference lay in a few minutes of wiggling.

The same working group then tested two further strategies in fifteen people. Three hours of sitting without preparation lowered flow mediated dilation from 3.8 to 1.5 per cent. If an hour of cycling preceded the sitting, the value stayed practically unchanged at 3.6 per cent. And three hours of standing instead of sitting preserved vascular function as well, at 4.3 instead of 4.1 per cent.

That is encouraging news for everyone who cycles in the morning or uses a standing desk. Both appear able to cushion this particular effect.

Healy and a team from Brisbane and Melbourne analysed accelerometer data from 168 adults in the Australian AusDiab study. They counted not only how long someone sat, but how often that sitting was interrupted. Independently of total sitting time and independently of moderate to vigorous activity, a higher number of interruptions was associated with more favourable values for waist circumference, BMI, triglycerides and the two hour glucose value.

For you that means: it is not only the sum that counts, the pattern counts too. This is a cross-sectional observation, it proves no cause. But it triggered the right experiments.

Dunstan and a team from the Baker IDI Heart and Diabetes Institute in Melbourne had nineteen overweight adults between 45 and 65 years old go through three daily routines. Sitting continuously, or walking lightly for 2 minutes every 20 minutes, or walking moderately for 2 minutes every 20 minutes. Afterwards there was a standardised drink with 75 grams of glucose and 50 grams of fat.

Over the following five hours the glucose area was 6.9 units with continuous sitting. With light walking breaks it was 5.2, with moderate ones 4.9. For insulin the difference was even clearer: 828.6 compared with 633.6 and 637.6. And it is remarkable that the light variant was practically on a par with the moderate one.

Larsen and the same Melbourne environment repeated the experiment over three consecutive days with seventeen two minute walking breaks a day. On the first and on the third day the glucose area with continuous sitting was 1.3 and 1.5 units higher respectively than with breaks. The difference therefore held over the days, but it did not grow. For triglycerides no difference was found.

For you that means: movement breaks are not an account that fills up. They are a daily business.

Dempsey and colleagues tested whether it has to be walking at all. Twenty-four overweight adults with type 2 diabetes went through three days of eight hours each: sitting continuously, 3 minutes of slow walking every 30 minutes, or 3 minutes of simple resistance exercises while standing every 30 minutes, meaning half squats, calf raises, glute contractions and knee raises.

Both movement variants came out clearly more favourable for glucose, insulin and C-peptide after a meal than continuous sitting, with very similar values. For triglycerides only the resistance variant showed a significant difference, 2.9 compared with 4.8 units. For you that means: if you sit in an open plan office or cannot take a lap while working from home, you still have an option.

Buffey and a team from the University of Limerick pooled seven randomised crossover studies that had tested standing breaks and light walking breaks directly against each other. Standing breaks lowered glucose after a meal measurably, with an effect size of minus 0.31. On insulin and systolic blood pressure they had no significant effect.

Light walking breaks lowered glucose with minus 0.72 and insulin with minus 0.83. And in the direct comparison, walking breaks came out more favourably than standing breaks as well. The authors therefore call light walking the superior form of break.

Duvivier and a team from Maastricht University had 24 overweight adults, on average 64 years old, live four days in each of two patterns. Once with 13.5 hours of sitting a day, once with only 7.6 hours of sitting alongside 4 hours of standing and 4.3 hours of light walking. The meals were standardised.

In the more movement rich variant the insulin area was 20 per cent lower and insulin sensitivity in the Matsuda index 16 per cent higher. Fasting triglycerides were 32 per cent lower, non-HDL cholesterol 7 per cent, apolipoprotein B 4 per cent, and HDL cholesterol was 7 per cent higher. Diastolic blood pressure was lower as well.

This is perhaps the study closest to everyday life in this field. What is notable is that no exercise took place here. Only everyday movement distributed differently.

Hamilton and colleagues, by then at the University of Houston, described in 2022 a movement performed while seated: the ball of the foot stays on the floor, the heel is repeatedly lifted and lowered again. They call it the soleus push-up. In their measurements the soleus was able to raise its local oxygen turnover strongly over hours without fatiguing, including in untrained participants. Muscle biopsies showed only minimal glycogen use in the process. So the muscle apparently drew on blood sugar and blood fats instead of on its own store.

The systemic effect was clear: a 52 per cent smaller rise in glucose after a sugar load, with a difference of roughly 50 milligrams per decilitre between the first and second hour, and a 60 per cent smaller rise in insulin. VLDL triglycerides in the blood also came out more favourably.

Chen and a team from Deakin University in Australia screened 1,778 titles and kept 15 cohort and case control studies in the end. Their conclusion was uncomfortable: in these data a sedentary lifestyle on its own was not established as a risk factor for the development of low back pain. Only a single high quality cohort study found an association, and exclusively for sitting at work.

Dzakpasu and a team from Melbourne took up the topic again in 2021, with 79 studies and separated by domain of life. In cross-sectional data they did find associations. Total sitting time across the day was associated with low back pain, with an odds ratio of 1.19. Self-reported sitting at work was at 1.47 for low back pain and at 1.73 for neck and shoulder pain. For pain in arms and legs no significant association was found.

The prospective evidence, however, stayed inconclusive. And the authors themselves stress that reverse causality cannot be ruled out: people who are in pain may sit more. The intervention part is interesting. Studies that specifically reduced sitting at work reported fewer low back, neck and shoulder complaints.