Sedentary behaviour and health in adults: an overview of systematic reviews

This is an overview of systematic reviews — a "review of reviews" — that synthesised findings from 18 separate systematic reviews. The researchers asked two questions:

1. **What is the relationship between different types of sedentary behaviour** (e.g., TV viewing, computer use, driving, total sitting time) and health outcomes in adults?

2. **What is the relationship between different patterns of sedentary behaviour** (e.g., prolonged sitting bouts vs. frequent breaks, total daily sitting time) and health outcomes in adults?

The health outcomes examined included:

**Brain and mental health:** cognitive function, depression, dementia, anxiety

**Physical function:** mobility, disability, frailty, pain

**Quality of life:** physical and mental health-related quality of life

**Body composition:** body fat, waist circumference, BMI

**Cardiometabolic markers:** blood pressure, blood glucose, cholesterol, triglycerides

**Other 24-hour movement behaviours:** physical activity levels, sleep quality

The comparator was either lower levels of sedentary behaviour or uninterrupted sitting (for intervention studies that broke up sitting time).

The 18 systematic reviews included data from hundreds of individual studies, collectively covering:

**Population:** Community-dwelling adults aged 18 years and older (including older adults aged 65+)

**Sample sizes per review:** Ranged from ~1,500 to over 100,000 participants depending on the review

**Settings:** General population samples from multiple countries (Canada, USA, UK, Australia, Japan, European nations)

**Health status:** Mostly healthy adults, but some reviews included adults with chronic conditions (e.g., type 2 diabetes, cardiovascular disease, osteoarthritis)

**Age range:** 18 to 100+ years, with several reviews specifically focused on older adults (65+)

The overview did not include children, adolescents, or institutionalised adults (e.g., hospitalised patients, nursing home residents).

The included systematic reviews used a mix of measurement approaches:

**Sedentary behaviour measurement:**

**Self-report questionnaires:** International Physical Activity Questionnaire (IPAQ), Sedentary Behaviour Questionnaire (SBQ), domain-specific questions (e.g., "How many hours per day do you watch TV?")

**Device-based measurement:** Accelerometers (hip-worn or thigh-worn) that measure posture and movement intensity; some studies used inclinometers (e.g., activPAL) that distinguish sitting from standing

**Pattern measurement:** Total daily sedentary time, number of sedentary bouts (e.g., bouts ≥20 minutes), number of breaks in sedentary time, average bout duration

**Health outcome measurement:**

**Cognitive function:** Standardised neuropsychological tests (e.g., Mini-Mental State Examination, Montreal Cognitive Assessment, processing speed tests, executive function tests)

**Depression:** Validated scales (e.g., Center for Epidemiologic Studies Depression Scale, Geriatric Depression Scale, Patient Health Questionnaire-9)

**Function and disability:** Self-reported mobility limitations, gait speed tests, chair stand tests, activities of daily living scales

**Quality of life:** SF-36 (Short Form Health Survey), EQ-5D, WHOQOL-BREF

**Body composition:** Dual-energy X-ray absorptiometry (DXA), bioelectrical impedance, waist circumference measured by tape, BMI calculated from measured height and weight

**Cardiometabolic markers:** Fasting blood glucose, HbA1c, triglycerides, HDL and LDL cholesterol, resting blood pressure (systolic and diastolic)

**Study design:** This is an **overview of systematic reviews** (also called a "umbrella review"). The authors searched five electronic databases (MEDLINE, EMBASE, CINAHL, PsycINFO, SPORTDiscus) up to May 2019, with a 10-year search limit (2009–2019). They included only systematic reviews that met pre-specified criteria for population (adults 18+), exposure (sedentary behaviour types and/or patterns), and outcomes (brain/mental health, function/disability, quality of life, body composition, cardiometabolic risk, other movement behaviours).

**Inclusion/exclusion:** Reviews had to be systematic (explicit search strategy, clear inclusion criteria). They could include observational studies (cross-sectional, cohort, case-control) or intervention studies (randomised controlled trials, quasi-experimental). Reviews focused only on mortality, cardiovascular disease, type 2 diabetes, or cancer were excluded because those outcomes were already covered by the 2018 US Physical Activity Guidelines Advisory Committee report.

**Quality assessment:** The authors did not re-rate the quality of individual studies within each review. Instead, they reported the quality ratings assigned by the original review authors. These ratings varied from "low" to "high" depending on the review and outcome. The overview authors then assessed the **certainty of evidence** across all reviews using a modified GRADE approach (Grading of Recommendations Assessment, Development and Evaluation). The overall certainty was rated as **low to very low** for most outcomes.

**Statistical approach:** Because this is an overview of reviews, the authors did not perform a new meta-analysis. They narratively synthesised findings across reviews, noting where results were consistent or inconsistent.

**What this design can and cannot prove:**

**Can prove:** This design provides a high-level summary of the best available evidence across multiple health outcomes. It can identify consistent patterns (e.g., "most reviews find a link between TV time and depression") and gaps (e.g., "no reviews examined pain outcomes").

**Cannot prove:** This design cannot establish causation. Most of the included reviews were based on observational studies (cross-sectional and cohort), which can show associations but cannot rule out reverse causation (e.g., people who are already depressed may watch more TV) or confounding (e.g., people who sit a lot may also have poor diets, smoke more, or exercise less). The intervention studies included were mostly short-term (hours to days) and focused on surrogate markers (e.g., blood glucose) rather than hard outcomes (e.g., heart attacks).

**Major methodological weakness:** The overview is limited by the quality of the underlying reviews. Many reviews had low-quality evidence due to reliance on self-reported sedentary behaviour (which is notoriously inaccurate), cross-sectional designs, and failure to adequately control for physical activity levels. The authors note that the certainty of evidence was "low to very low" for most conclusions.

**Total sedentary behaviour and TV viewing (most consistent findings):**

**Cognitive function:** 4 of 4 reviews found that higher total sedentary time was associated with **worse cognitive function** in older adults. However, the relationship varied by type: TV viewing was consistently linked to worse cognition, while computer/Internet use was associated with **better** cognitive function in older adults (possibly because these activities are mentally stimulating).

**Depression:** 4 of 5 reviews found that higher sedentary behaviour (especially TV viewing) was associated with **greater risk of depression**. One review reported a dose-response: each additional hour of daily sitting was associated with a 5–10% increase in depression risk (exact numbers varied by study).

**Function and disability:** 2 of 2 reviews found that higher sedentary time was associated with **worse physical function** and greater disability in older adults. One review reported that older adults in the highest quartile of sedentary time had 1.5–2.0 times higher odds of mobility limitations compared to those in the lowest quartile.

**Quality of life:** 2 of 2 reviews found that higher sedentary behaviour was associated with **lower physical health-related quality of life**. The relationship with mental health-related quality of life was inconsistent.

**Body composition:** Intervention studies that reduced or broke up sedentary time showed **small improvements in body composition** (waist circumference reduced by ~0.5–1.5 cm, body fat reduced by ~0.3–0.8 kg) over periods of 4–24 weeks.

**Cardiometabolic markers:** Intervention studies that broke up prolonged sitting with short movement breaks (e.g., 2–5 minutes of light walking every 30–60 minutes) showed **acute reductions in post-meal blood glucose and insulin** (glucose reduced by ~10–20% compared to uninterrupted sitting). These effects were observed in single-day lab experiments.

**Physical activity:** 2 of 2 reviews found that higher sedentary behaviour was associated with **lower physical activity levels** — but this is partly a measurement artefact (more time sitting leaves less time for activity).

**Patterns of sedentary behaviour (bouts vs. breaks):**

**Breaks in sedentary time:** Intervention studies consistently found that breaking up prolonged sitting with short, frequent movement breaks (e.g., 2 minutes of walking every 30 minutes) **acutely reduced post-meal blood glucose and insulin** compared to uninterrupted sitting. The effect was larger in people with higher baseline glucose (e.g., those with type 2 diabetes or prediabetes).

**Sedentary bouts:** Longer uninterrupted sitting bouts were associated with worse cardiometabolic markers in observational studies, but the evidence was limited and inconsistent.

**Timing of sedentary behaviour:** No reviews examined whether sedentary behaviour in the morning vs. evening had different health effects.

**Type-specific findings:**

**TV viewing:** Most consistently associated with negative health outcomes (depression, worse cognition, obesity, poor quality of life). This may be partly because TV viewing is often accompanied by unhealthy snacking and is a passive, non-stimulating activity.

**Computer/Internet use:** Mixed findings. In older adults, computer use was associated with **better** cognitive function (possibly due to mental stimulation). In younger adults, excessive computer use was associated with worse mental health in some studies.

**Occupational sitting:** Limited evidence. Some studies found links between prolonged occupational sitting and worse cardiometabolic health, but findings were inconsistent.

**Driving:** Very limited evidence. One review found that driving time was associated with higher BMI, but only in men.

**Effect sizes (where reported):**

**Depression:** Odds ratios ranged from 1.10 to 1.50 for high vs. low sedentary behaviour (i.e., 10–50% higher odds of depression)

**Cognitive decline:** Hazard ratios of 1.20–1.40 for high vs. low sedentary time over 5–10 year follow-up

**Waist circumference:** Mean reduction of 0.8–1.5 cm in intervention studies lasting 8–24 weeks

**Post-meal glucose:** 10–20% reduction with breaking up sitting vs. uninterrupted sitting in acute lab studies

**Mobility limitations:** Odds ratios of 1.5–2.0 for highest vs. lowest sedentary time in older adults

**In plain English:**

**Depression:** If you sit 9–10 hours per day, your odds of being depressed are roughly 10–50% higher than someone who sits 4–5 hours per day. This is a meaningful increase — comparable to the effect of being physically inactive or having poor sleep.

**Cognitive function:** Older adults who sit the most have about 20–40% higher risk of cognitive decline over 5–10 years compared to those who sit the least. However, this is partly offset by the type of sitting: using a computer appears to be protective, while watching TV is harmful.

**Body composition:** Replacing 30–60 minutes of sitting with standing or light walking each day might reduce waist circumference by about 1 cm over 3–6 months. That's roughly the equivalent of losing 1–2 kg of body fat — noticeable but modest.

**Blood sugar:** Breaking up sitting with 2–5 minutes of walking every 30 minutes can reduce the spike in blood sugar after a meal by 10–20%. For someone with prediabetes, this could be the difference between a normal and abnormal glucose reading after lunch.

**Physical function:** Older adults who sit the most are about 50–100% more likely to have trouble walking or climbing stairs. This is a large effect — comparable to the difference between being a regular exerciser vs. being completely sedentary.

**Important caveat:** These are associations, not proven causes. The actual effect of reducing sitting might be smaller (or larger) once other factors like diet, exercise, and genetics are accounted for.

**What the authors acknowledge:**

**Low certainty of evidence:** The overall certainty was rated as "low to very low" for most outcomes. This means future research is very likely to change the conclusions.

**Reliance on self-report:** Most studies used self-reported sedentary behaviour, which is notoriously inaccurate. People tend to underestimate their sitting time by 1–2 hours per day.

**Cross-sectional designs:** Many studies measured sitting and health at the same time, making it impossible to know which came first. Does sitting cause depression, or do depressed people sit more?

**Confounding:** Few studies adequately controlled for physical activity levels, diet, smoking, or socioeconomic status. People who sit a lot may also have other unhealthy habits.

**Publication bias:** Studies finding no link between sitting and health may be less likely to be published.

**Limited intervention data:** Most intervention studies were short-term (hours to days) and measured surrogate markers (blood glucose, insulin) rather than hard outcomes (heart attacks, strokes, death).

**What a critical reader would add:**

**No blinding:** In intervention studies, participants knew whether they were sitting or breaking up sitting. This could affect behaviour (e.g., eating less when walking).

**Industry funding:** Some studies on standing desks were funded by standing desk manufacturers. The overview did not assess funding sources.

**Population limits:** Most studies were in high-income countries (Canada, USA, UK, Australia). Results may not apply to low-income settings where sedentary behaviour patterns differ.

**Age range:** The overview combined adults 18–64 and older adults 65+. The relationship between sitting and health may differ by age (e.g., computer use may be beneficial for older adults but harmful for younger adults).

**No dose-response analysis:** The overview did not quantify the optimal amount of sitting or the minimum number of breaks needed for health benefits.

**Missing outcomes:** Pain, productivity, and sleep were listed as outcomes of interest but no reviews were found that examined these.

For someone running their own n=1 experiment:

### What to test

**Primary intervention:** Replace 30–60 minutes of daily sitting with standing or light walking, OR break up prolonged sitting with 2–5 minute movement breaks every 30–60 minutes.

**Specific dose:** Try either (a) standing desk for 2–3 hours per workday, or (b) a timer that prompts you to walk for 2 minutes every 30 minutes during work hours.

**Type matters:** If you're an older adult, consider replacing TV time with computer/Internet use (e.g., online puzzles, learning a new skill) — this may benefit cognition while still being sedentary.

### Minimum meaningful duration

**For blood sugar effects:** You can see acute changes within a single day. Measure your glucose 1–2 hours after a meal on a sitting day vs. a break-up-sitting day.

**For body composition:** Run the experiment for at least 8–12 weeks. Changes in waist circumference or body fat are unlikely to be detectable before 4–6 weeks.

**For mood/cognition:** 2–4 weeks may be enough to notice subjective changes in energy, focus, or mood. Use daily ratings.

### What to measure (specific metrics)

**Primary outcome:** Choose ONE of the following based on your goal:

- **Blood sugar:** Fasting glucose (morning, before eating) OR post-meal glucose (1 hour after a standardised meal). Use a home glucometer.

- **Body composition:** Waist circumference (measured at navel level, after exhaling, same time of day) OR body weight (morning, after bathroom, before eating).

- **Mood:** Daily rating on a 1–10 scale for "dep