Replacing sitting time with standing or stepping: associations with cardio-metabolic risk biomarkers

The researchers tested whether swapping 2 hours of daily sitting time with either standing or stepping (walking around) was associated with better cardio-metabolic health markers. They did not actually make people change their behaviour — instead, they used a statistical technique called isotemporal substitution to model what would happen if someone replaced sitting with standing or stepping, based on the actual behaviour patterns of nearly 700 adults.

The outcome measures were:

• Body composition: Body mass index (BMI, kg/m²) and waist circumference (cm)
• Blood pressure: Systolic and diastolic (mmHg)
• Blood sugar control: Fasting plasma glucose (mmol/L), 2-hour plasma glucose after an oral glucose tolerance test (mmol/L), and HbA1c (a marker of average blood sugar over ~3 months, %)
• Blood fats (lipids): HDL-cholesterol ("good" cholesterol, mmol/L), LDL-cholesterol ("bad" cholesterol, mmol/L), total/HDL-cholesterol ratio, and triglycerides (mmol/L)

• Sample size: 698 adults with complete data
• Population: Participants from the 2011/12 Australian Diabetes, Obesity, and Lifestyle Study (AusDiab), a national survey
• Age range: 36–80 years (mean 57.9 years, SD 9.9 years)
• Sex: 57% women
• Setting: Community-dwelling adults across Australia
• Health status: Included people with and without diabetes, obesity, and other metabolic conditions — this was a general population sample, not a clinical trial of healthy volunteers

• Sitting, standing, and stepping time: Measured using the activPAL3 monitor, a small device worn on the thigh that uses accelerometry and posture detection to distinguish sitting/lying from standing from stepping. Participants wore it for 7 consecutive days during waking hours. This is considered the gold standard for posture-based measurement because it can tell the difference between sitting and standing (unlike hip-worn accelerometers).
• Cardio-metabolic biomarkers: Measured via standard clinical protocols:
  • BMI from measured height and weight
  • Waist circumference from a tape measure at the narrowest point
  • Blood pressure from an automated monitor after 5 minutes seated rest
  • Fasting blood samples (≥8 hours fast) analysed for glucose, HbA1c, and lipids
  • 2-hour plasma glucose from a 75g oral glucose tolerance test (OGTT)
• Confounders: Self-reported via questionnaire — age, sex, education, smoking, alcohol, diet quality, total physical activity (not just stepping), and medication use for diabetes, hypertension, and cholesterol

Study design: Cross-sectional observational study using data from a single wave of the AusDiab cohort.

What they did: Participants wore the activPAL3 monitor for 7 days. Researchers then calculated each person's average daily time spent sitting, standing, and stepping. They used a statistical technique called isotemporal substitution analysis, which models the effect of replacing a fixed amount of time (here, 2 hours/day) from one behaviour (sitting) with another (standing or stepping), while keeping total waking time constant. This is like asking: "If two people have the same total waking hours, but one sits 2 hours less and stands 2 hours more, what's the difference in their health markers?"

Why that design matters:

• Cross-sectional design: This measures all variables at one point in time. It can show associations but cannot prove causation. You cannot tell whether sitting less caused better health, or whether healthier people simply choose to sit less.
• No randomisation: Participants were not assigned to sit, stand, or step. The analysis adjusts for measured confounders (age, diet, etc.), but unmeasured confounders (e.g., genetics, job type, chronic pain) could explain the results.
• No blinding: Not applicable in an observational study, but participants knew they were wearing an activity monitor, which could alter behaviour (Hawthorne effect).
• Duration: 7 days of monitoring gives a reasonable snapshot of habitual behaviour, but one week may not capture seasonal or weekly variation.
• Statistical approach: Isotemporal substitution is a powerful modelling technique that accounts for the fact that time is finite — more time in one behaviour means less in another. However, it assumes the relationship is linear and that the effect of replacing 2 hours is the same regardless of baseline sitting time, which may not be true.

What this design can prove:

• That sitting, standing, and stepping time are associated with cardio-metabolic biomarkers after adjusting for many confounders.
• That the direction of association (more standing/stepping = better markers) is consistent with the hypothesis that replacing sitting is beneficial.

What this design cannot prove:

• That changing sitting time causes changes in health markers.
• That the modelled effects would actually occur if someone started standing more tomorrow.
• That the effects are the same for all individuals (e.g., young vs. old, healthy vs. diabetic).

Major methodological weaknesses:

• Cross-sectional design — cannot establish temporality or causation
• Self-reported confounders (diet, alcohol, medication) are prone to recall bias
• No data on sleep duration or quality, which is a major confounder for metabolic health
• The sample is predominantly older (mean 58 years) and Australian — results may not generalise to younger populations or other ethnicities
• The activPAL3 cannot distinguish standing still from standing while fidgeting or shifting weight, which may matter metabolically

Primary outcomes (all reallocations of 2 hours/day):

Sitting-to-standing reallocations (replace 2 hours sitting with 2 hours standing):

• Fasting plasma glucose: ~2% lower (p < 0.05)
• Triglycerides: ~11% lower (p < 0.05)
• Total/HDL-cholesterol ratio: ~6% lower (p < 0.05)
• HDL-cholesterol: 0.06 mmol/L higher (p < 0.05)
• BMI, waist circumference, blood pressure, HbA1c, 2-hour glucose, LDL-cholesterol: No significant association

Sitting-to-stepping reallocations (replace 2 hours sitting with 2 hours walking):

• BMI: ~11% lower (p < 0.05)
• Waist circumference: ~7.5 cm lower (p < 0.05)
• 2-hour plasma glucose: ~11% lower (p < 0.05)
• Triglycerides: ~14% lower (p < 0.05)
• HDL-cholesterol: 0.10 mmol/L higher (p < 0.05)
• Fasting glucose, HbA1c, blood pressure, LDL-cholesterol, total/HDL ratio: No significant association

Standing-to-stepping reallocations (replace 2 hours standing with 2 hours walking):

• BMI: ~10% lower (p < 0.05)
• Waist circumference: ~7 cm lower (p < 0.05)
• 2-hour plasma glucose: ~11% lower (p < 0.05)
• All other markers: No significant association

Key non-significant findings:

• Blood pressure (systolic and diastolic) was not significantly associated with any reallocation
• HbA1c was not significantly associated with any reallocation
• LDL-cholesterol was not significantly associated with any reallocation

Important note on interpretation: These are cross-sectional associations, not experimental effects. The "11% lower BMI" for sitting-to-stepping means that people who naturally walk 2 hours more and sit 2 hours less have ~11% lower BMI on average — not that you will lose 11% of your body weight by walking 2 hours/day.

To put these numbers in perspective:

• Standing instead of sitting for 2 hours/day was associated with triglyceride levels about 11% lower. For someone with triglycerides of 1.7 mmol/L (borderline high), that would be a drop to ~1.5 mmol/L — a modest but clinically meaningful improvement. The 2% lower fasting glucose is tiny: for someone with fasting glucose of 5.5 mmol/L, that's about 0.11 mmol/L, roughly the effect of skipping one sugary drink.

• Stepping instead of sitting for 2 hours/day was associated with a 7.5 cm smaller waist circumference. That's roughly 3 inches — a large difference. But remember: this is comparing people who habitually walk 2 hours/day more versus those who sit 2 hours/day more. The difference likely reflects years of accumulated behaviour, not a quick fix.

• The HDL-cholesterol increases (0.06–0.10 mmol/L) are modest. A typical target for HDL is >1.0 mmol/L in men and >1.3 mmol/L in women. An increase of 0.10 mmol/L is roughly what you might get from switching from a sedentary to a moderately active lifestyle over several months.

• The lack of effect on blood pressure and HbA1c is notable. These are harder to shift with movement alone, especially in a general population where many people already have normal values.

What the authors acknowledge:

• Cross-sectional design prevents causal inference
• Isotemporal substitution assumes linear relationships
• Possible residual confounding from unmeasured variables (e.g., sleep, stress, occupational sitting patterns)
• The activPAL3 cannot detect non-ambulatory standing activities (e.g., standing while fidgeting, shifting weight, or doing upper-body work)
• The sample is predominantly older and Australian, limiting generalisability

What a critical reader would note:

• No adjustment for sleep: Sleep duration and quality are strongly linked to metabolic health and also to activity patterns. People who sit less and stand more may also sleep differently, which could confound results.
• Self-reported confounders: Diet quality, alcohol intake, and medication use were self-reported, which introduces measurement error. People who stand more may also underreport unhealthy behaviours.
• Multiple comparisons: The study tested many outcomes (12 biomarkers × 3 reallocation types = 36 comparisons). Some significant results may be due to chance. The authors did not adjust for multiple comparisons.
• Effect sizes are cross-sectional, not longitudinal: The 11% lower BMI for stepping is not a prediction of weight loss — it's a comparison between groups. People who walk 2 hours/day may also eat differently, have different jobs, or have different genetics.
• No data on standing quality: Standing still vs. standing while shifting weight, pacing, or doing light activity may have very different metabolic effects. The activPAL3 treats all standing as equivalent.
• Industry funding: The study was funded by Australian government health agencies (NHMRC, ARC), so no obvious industry bias. However, the standing desk industry has a vested interest in promoting standing, and this paper is often cited by them.

For someone running their own n=1 experiment:

What to test

• Primary test: Replace 2 hours of daily sitting with standing (e.g., using a standing desk, standing during phone calls, standing while watching TV). This is the easiest intervention and the one with the lowest barrier.
• Secondary test: Replace 2 hours of daily sitting with stepping (e.g., walking meetings, treadmill desk, walking during lunch, pacing while on calls). This is more metabolically demanding but likely has larger effects.
• Tertiary test: Replace 2 hours of standing with stepping (e.g., instead of standing at your desk, walk around the block). This tests whether movement adds benefit beyond just being upright.

Minimum meaningful duration

• For blood sugar and triglyceride changes: 2–4 weeks. These markers can shift relatively quickly with changes in physical activity and posture.
• For waist circumference and BMI: 8–12 weeks minimum. Body composition changes require sustained behaviour change.
• For HbA1c: 3 months minimum. HbA1c reflects average blood sugar over ~3 months.
• For blood pressure: 4–6 weeks. Blood pressure can respond to activity changes within weeks, but is also highly variable day-to-day.

What to measure

• Primary metrics (weekly):
  • Fasting glucose (mmol/L) — measure at the same time each morning, after ≥8 hours fast
  • Triglycerides (mmol/L) — from a fasting blood test; can use a home lipid panel or lab test
  • Waist circumference (cm) — measure at the narrowest point, after exhaling, using a consistent tape measure
  • Body weight (kg) — same time each morning, after voiding, before eating/drinking
• Secondary metrics (biweekly or monthly):
  • Blood pressure (systolic/diastolic, mmHg) — measure at the same time of day, after 5 minutes seated rest, using a validated home monitor
  • HDL-cholesterol (mmol/L) — from a fasting blood test
  • 2-hour post-meal glucose (mmol/L) — if you have a glucose monitor, test 2 hours after a standardised meal
• Behaviour tracking:
  • Daily sitting, standing, and stepping time — use a posture-based monitor (activPAL is expensive; alternatives include the Apple Watch with stand reminders, or a simple log of "standing hours per day")
  • Sleep duration and quality — use a sleep tracker or diary, as sleep confounds metabolic health

Key confounds to control for

• Diet: Keep your diet as consistent as possible during the experiment. Track what you eat (even roughly) to ensure you're not unconsciously eating more or less when you stand more.
• Sleep: Aim for consistent sleep duration (7–9 hours) and timing. Poor sleep can raise glucose and triglycerides independently of activity.
• Stress: Track perceived stress (e.g., on a 1–10 scale daily). Stress hormones raise blood sugar and blood pressure.
• Medication: If you take medication for diabetes, hypertension, or cholesterol, do not change your dose during the experiment without medical supervision. Note any changes.
• Time of year: Seasonal changes in daylight, temperature, and activity patterns can affect mood and metabolism. Run your experiment in a consistent season or control for this.
• Caffeine and alcohol: Keep these consistent. Caffeine can raise blood pressure and glucose acutely; alcohol affects triglycerides and sleep.

What a positive result would look like

• For standing intervention: After 4 weeks of replacing 2 hours sitting with standing, you see:
  • Fasting glucose decrease by 0.1–0.2 mmol/L (2–4% drop)
  • Triglycerides decrease by 0.2–0.4 mmol/L (10–15% drop)
  • HDL-cholesterol increase by 0.05–0.10 mmol/L
  • Waist circumference decrease by 1–2 cm (if sustained for 8+ weeks)
• For stepping intervention: After 8 weeks of replacing 2 hours sitting with walking, you see:
  • Waist circumference decrease by 2–5 cm
  • Body weight decrease by 1–3 kg
  • 2-hour post-meal glucose decrease by 0.5–1.0 mmol/L
  • Triglycerides decrease by 0.3–0.6 mmol/L
• No change in: Blood pressure, HbA1c, or LDL-cholesterol within 4–8 weeks is expected and not a failure. These markers are harder to shift.
• What to watch for: If you see no change after 8 weeks of consistent standing, consider:
  • Are you actually standing for