Body phenotyping

BMI misses the normal-weight person carrying dangerous visceral fat — thin outside, fat inside (TOFI) — and it falsely flags the person with obesity whose fat distribution is otherwise benign. Among adults BMI classified as normal weight, 18% had an elevated waist-to-hip ratio: central adiposity BMI did not account for. Among those classified as having obesity, about 45% had a low waist-to-hip ratio — BMI called them high risk, and their fat distribution did not agree.

WHAT ELEVATED CENTRAL ADIPOSITY CARRIES. For adults of normal weight or overweight, an elevated waist circumference or waist-to-hip ratio carried 15% to 50% greater risk across the cardiovascular outcomes studied. The largest relative increases were for all-cause death and cardiovascular disease.

Attributable burden is a different question from relative risk. Among people WITH OBESITY, 49% of heart failure events, 46% of atrial fibrillation events and 36% of coronary heart disease mortality were attributable to elevated waist circumference.

WHEN BMI AND THE TAPE DISAGREE. Among people with obesity, a low WAIST CIRCUMFERENCE was not simply neutral: it translated into lower all-cause mortality, with no other outcome differing materially.

The sex difference is a WAIST-TO-HIP finding, not a waist-circumference one — the metric changes here, and the change matters. Men with obesity and a low waist-to-hip ratio were not risk-free: atrial fibrillation remained elevated compared with men who had both normal weight and a low ratio. Women with obesity and a low waist-to-hip ratio remained at higher risk across all cardiovascular outcomes than normal-weight women with a low ratio.

WHAT THIS DOES NOT SAY. Dardari and colleagues report observational data. It shows that waist circumference and waist-to-hip ratio reclassify cardiovascular risk beyond BMI. It does not show that deliberately reducing either causally lowers cardiovascular events; residual confounding cannot be excluded, and no cause-and-effect relationship can be established from this design. The biology makes it plausible; the study does not make it proven.

MEASURING IT THE SAME WAY EVERY TIME. Small variations in technique move someone between risk categories, so standardisation matters more here than in almost any other measurement in this app. Waist: stand with feet together, and pass a non-stretch tape around the abdomen over the top of the iliac crest, snug without compressing. Read at the end of a normal breath out, and use the same landmark for every follow-up. Hip: stand with feet together, and pass the tape around the widest girth at the level of the buttocks.

THE TWO RATIOS. Waist-to-height ratio is waist ÷ height. Waist-to-hip ratio is waist ÷ hip. Both sides in the same unit — centimetres or inches, but not one of each. Verve leads with waist-to-height because it needs only the waist and the height already on record, so it appears the moment a waist is measured; waist-to-hip needs a second tape measurement nothing can supply automatically, and is offered as the complementary reading. Both are shown on the waist card itself, beside the tape that produced them.

MEASURING YOUR HIP. Stand with your feet together. Pass a non-stretch tape around the widest girth at the level of the buttocks, level all the way around and snug without compressing. There is no reference range for a hip on its own and Verve does not offer one — a hip means something only as the second half of the ratio.

THE CUT-OFFS BELOW ARE THE WHO AND IDF CONSENSUS VALUES, NOT THE PAPER'S. Dardari and colleagues defined high waist and high waist-to-hip ratio within their own cohorts; the thresholds Verve judges your measurement against are the standard clinical ones. Waist circumference: 102 cm (40 in) men, 88 cm (35 in) women in the general population; 90 cm (35 in) men, 80 cm (31.5 in) women in Asian populations. Waist-to-hip: 0.90 men, 0.85 women. Waist-to-height: 0.50 for everyone. Only waist circumference has ethnicity-specific cut-offs — waist-to-hip and waist-to-height do not, and no guideline publishes ancestry-adjusted versions of them.

Verve applies the IDF metabolic-syndrome waist cut-off (94 cm men, 90 cm men of Asian ancestry, 80 cm women) inside the metabolic-syndrome tally, because that is the number the Alberti 2009 criteria are written against. The higher WHO figures above describe substantially increased cardiovascular risk. They are two questions, and they have two published answers.

EWGSOP2 does not define a cut-off for FFMI — fat-free mass index is not part of its algorithm at all. Its muscle-mass criterion is appendicular lean mass (ALM) and its height-indexed form, ALMI, measured by DXA. Appendicular means the limbs: EWGSOP2 isolates arm and leg lean mass because it reflects the skeletal muscle that carries strength and mobility, excluding trunk lean mass, which includes organs.

The algorithm is sequential, and the order is the point: STRENGTH DEFINES IT, MASS CONFIRMS IT, PERFORMANCE GRADES IT. Low grip strength identifies PROBABLE sarcopenia; low ALM or ALMI CONFIRMS it; and low physical performance — gait speed or Timed Up and Go — grades it as SEVERE. Low muscle mass on its own is not sarcopenia under EWGSOP2, which is why Verve never reports it as such.

EWGSOP2 thresholds. Grip strength: <27 kg men, <16 kg women — defines probable sarcopenia. ALM absolute: <20 kg men, <15 kg women — the alternative mass threshold, used when height is not on file. ALMI (ALM ÷ height², DXA): <7.0 kg/m² men, <5.5 kg/m² women — confirms. Gait speed (4 m usual pace): ≤0.8 m/s — severity. Timed Up and Go: ≥20 s — alternative to gait speed.

The women's ALMI cut-off differs by guideline AND by measurement, and the two must not be mixed: EWGSOP2 reads <5.5 kg/m² by DXA, while AWGS reads <5.4 kg/m² by DXA and <5.7 kg/m² by BIA. Verve applies the DXA figure, because ALM is entered from a DXA report.

FFMI cut-offs exist within NUTRITION frameworks, not within the muscle-strength ones. ESPEN and GLIM both define them; EWGSOP2's sarcopenia algorithm does not use FFMI at all, relying on ALM and ALMI instead.

ESPEN 2015 consensus criteria for malnutrition (DXA-derived FFMI): low is <17 kg/m² in men and <15 kg/m² in women. FFMI is used only as an ALTERNATIVE to reduced BMI, and only once unintentional weight loss is present — more than 10% at any time, or more than 5% over three months. Neither FFMI nor BMI alone diagnoses malnutrition.

GLIM 2019 uses the same FFMI thresholds as its reduced-muscle-mass phenotypic criterion — <17 kg/m² men, <15 kg/m² women, equivalent to ALMI <7.0 and <5.5 per EWGSOP2. A diagnosis requires that phenotypic criterion PLUS at least one etiologic criterion: reduced intake or assimilation, or disease-related inflammation.

These cut-offs were validated by DXA in the SarcoPhAge cohort and were associated with increased mortality in malnourished older adults.

Separately, population screening studies using BIA-derived FFMI have proposed surrogate thresholds of 17.5–18 kg/m² in men and 14.4–15 kg/m² in women, correlating with DXA-defined low ALMI in Japanese cohorts, with one linking low FFMI to increased fall risk. These are research-derived and ethnicity-specific — not formal guideline cut-offs — so Verve classifies against ESPEN/GLIM and names them here rather than applying them silently.

SPPB stands for the SHORT PHYSICAL PERFORMANCE BATTERY. It was developed for the US National Institute on Aging's EPESE studies and published by Guralnik and colleagues in 1994, and it is the most widely used measure of lower-body function in older adults. Three tests — standing balance, walking speed, and rising from a chair — each score 0 to 4, and the three add to a total out of 12.

It matters because it predicts things a single measurement does not. Lower SPPB scores are associated with disability, hospitalisation, nursing-home admission and mortality, and EWGSOP2 uses the battery as one of its measures of physical performance — the step that grades sarcopenia as SEVERE once low strength and low muscle mass have already been established.

HOW TO DO IT — 1. BALANCE. Three stands, each held up to 10 seconds, done in order and stopped as soon as one is failed. Feet side by side, touching. Then semi-tandem: the heel of one foot beside the big toe of the other. Then full tandem: one heel directly in front of the other toes, as though on a line. Stand near a wall or a helper, arms free, and stop the clock the moment you move your feet or grab support. Score: 0 if side-by-side is not held for 10 s; 1 if side-by-side is held but semi-tandem is not; 2 if tandem is held under 3 s; 3 if tandem is held 3 to 9 seconds; 4 if tandem is held the full 10 s.

HOW TO DO IT — 2. WALKING SPEED. Mark a straight 4-metre course with clear space to start and stop. Walk it at your USUAL pace, not your fastest — this is the pace you would use crossing a room, and walking fast is the commonest way to get an unrepresentative result. Use a walking aid if you normally use one. Time from first movement to when the leading foot crosses the 4 m line, take the better of two attempts, and enter the speed in metres per second (4 divided by your time in seconds). Score: 4 if it takes under 4.82 s, 3 up to 6.20 s, 2 up to 8.70 s, 1 if slower.

HOW TO DO IT — 3. CHAIR STAND. Use a straight-backed chair against a wall. Fold your arms across your chest — no pushing off with hands, which is what makes this a leg test — and stand fully upright and sit down five times as fast as you safely can. Time all five repetitions. If you cannot do one without using your arms, the score is 0. Score: 4 under 11.2 s, 3 up to 13.6 s, 2 up to 16.6 s, 1 up to 60 s, 0 beyond that.

READING THE TOTAL. 10 to 12 is good function. 7 to 9 is intermediate and is where the risk curve starts to bend. 6 or below is poor and is the range most strongly associated with subsequent disability. A change of about 1 point is generally taken as the smallest meaningful change, so the number is worth repeating every few months rather than watching week to week.

SAFETY. Every part is optional and every part can be stopped. Do the balance stands within reach of something solid, and skip any test that feels unsafe on the day — a missing score is honest, and a fall is not worth a number.