Sitting 6 or More Hours a Day Was Linked to 45% Higher Death Risk in 7,333 Adults

Six or more hours of daily sitting time was linked to a 45 percent higher rate of death from any cause among US adults living with diabetes or prediabetes. The same analysis found that sitting time and depressive symptoms each carried risk on their own, and that people reporting both carried the steepest hazard of all.

Sitting Time has moved from lifestyle footnote to measured risk factor, and a new analysis in BMC Public Health, published 4 November 2026, attaches numbers to it for one of the groups with the most at stake. Among 7,333 American adults with diagnosed diabetes or prediabetes, those who reported sitting at least six hours a day died at a higher rate than those who sat less.

The work was led by Zhi Jin of the Department of Neurology at Shanghai Fifth People’s Hospital, Fudan University, with Yukun Zhang and Yuqing Li of the Department of Cardiology at the Second Hospital of Tianjin Medical University. It is a secondary analysis, not a new trial, and the authors are careful to frame it that way.

The team pooled six cycles of the National Health and Nutrition Examination Survey covering 2007 through 2018, then linked participants to national mortality records through December 31, 2019. The analytic sample held 5,141 adults with diabetes and 2,192 with prediabetes. Across the follow-up window, 1,197 people died of any cause and 326 died of heart disease.

Exposures were deliberately simple. Daily sitting time was split at a single threshold, under six hours or six hours and above. Depressive symptoms were defined by a Patient Health Questionnaire-9 score of 10 or higher. Survey-weighted Cox models then estimated hazard ratios for each exposure alone and for four combined patterns.

Why Sitting Time Carried Risk on Its Own

The cleanest result is the one for sitting alone. Adults reporting at least six hours a day had a hazard ratio of 1.45 for all-cause mortality, with a 95% confidence interval running from 1.25 to 1.69. Because that interval sits well clear of 1.0, the association is unlikely to be a statistical accident in this sample.

For heart disease mortality specifically, longer sitting time produced a hazard ratio of 1.57, with an interval of 1.27 to 1.95. Both estimates point the same direction, and the cardiac figure is slightly steeper than the all-cause figure.

These are relative hazards, not absolute risks. A ratio of 1.45 describes how two groups in this cohort compared with one another over the follow-up period. It does not translate into a personal probability for any individual reader.

Depressive Symptoms Carried a Separate Signal

Participants scoring 10 or higher on the PHQ-9 had a hazard ratio of 1.33 for all-cause mortality, with an interval of 1.06 to 1.67. For heart disease mortality the ratio was 1.62, with a wider interval of 1.06 to 2.48.

That wider interval matters. Only 326 heart disease deaths occurred in the whole sample, so estimates for that outcome rest on fewer events and carry more uncertainty than the all-cause numbers. The direction is consistent; the precision is lower.

The Combined Pattern Showed the Steepest Hazard

The authors built four joint-exposure patterns. Pattern 1 covered a PHQ-9 under 10 with sitting under six hours. Pattern 2 paired a high PHQ-9 with shorter sitting time. Pattern 3 paired a low PHQ-9 with longer sitting time. Pattern 4 combined a PHQ-9 of 10 or higher with six or more hours of sitting.

Measured against Pattern 1, Pattern 4 carried a hazard ratio of 1.88 for all-cause mortality, with an interval of 1.44 to 2.46. For heart disease mortality the ratio reached 2.46, with an interval of 1.57 to 3.85.

Why the Absence of Interaction Changes the Reading

Here the paper does something unusual for a risk-factor study: it argues against the most dramatic interpretation of its own numbers. The authors tested for both multiplicative and additive interaction and found no evidence of either.

In plain terms, the two exposures stacked but did not multiply. Long sitting time did not appear to amplify the association between depressive symptoms and death; it simply added its own risk alongside. The authors write that the absence of statistical interaction does not support interpreting sitting time as a modifier of the depression-mortality association.

That distinction is easy to lose in a headline and worth keeping. Two independent contributors are a different problem from one factor that supercharges another, and they call for different responses.

What the Design Can and Cannot Establish

NHANES is a cross-sectional survey linked forward to death records. That architecture is strong for describing associations across a nationally sampled population and weak for establishing cause. Sitting time was reported by participants at a single point, then treated as a stable exposure for years afterward.

Reverse causation is the obvious worry. Adults who were already frail, in pain, or managing advanced complications may have sat longer because of their condition rather than developing the condition because they sat. An observational design cannot separate those paths, which is why the authors call for prospective and intervention work.

What This Means for Readers Who Already Train

A regular gym habit does not automatically solve a long sitting day, and that is not a new idea. FLM has covered a controlled study in which prolonged sitting impaired leg blood vessel function even after twelve weeks of high-intensity interval training, which suggests the two behaviors occupy partly separate ground.

The population here matters too. Diabetes and prediabetes change the stakes of a sedentary day, and exercise research in that group has produced some of the strongest results anywhere, including a trial where a low-energy diet plus exercise brought diabetes remission in 54 percent of 96 adults.

Nothing in this analysis sets a target for anyone. The six-hour line was an analytical cut point chosen by the researchers, not a clinical threshold, and this study tested no intervention. What follows is general information rather than individualized advice, and anyone managing diabetes or prediabetes should take decisions with their own clinician.

Reasonable, cautious reading of the evidence base around sitting time looks something like this:

• Treat total daily sitting time as a quantity worth knowing, separately from how much you train.
• Interrupting long seated stretches is the behavior that trial evidence has most often examined.
• Volume still counts: FLM has reported that 12 minutes of vigorous activity a week matched 150 moderate minutes for lower mortality in accelerometer data.
• Occupation shapes the picture, and higher cardiometabolic risk was linked to both desk and physical jobs in 1,233 workers, so job type alone is a poor proxy.
• Mood is part of the clinical picture here, and anyone whose low mood is persistent deserves professional support rather than a movement plan alone.

Where Sitting Time Research Goes Next

The useful next step is the one the authors name. Randomized work that assigns people to reduce sitting time, then follows hard outcomes, would test whether the association seen here reflects something modifiable. Device-measured exposure would also sharpen a question that self-report blurs.

Until then this analysis does what large cohort work does well. It identifies a pattern across 7,333 people, quantifies it with intervals, and tells readers plainly how confident they should be. Reported sitting time of six hours or more travelled with higher mortality in this group, and depressive symptoms travelled with higher mortality too.