Body shape index calculator (ABSI)

The formula

The authors looked at how waist size grows with height and weight across US adults and found that, roughly, waist rises with weight to the power 2/3 and falls with height to the power 5/6 (Krakauer 2012). ABSI is your actual waist divided by that expected waist (Krakauer 2012):

ABSI = waist ÷ (BMI^2/3 × height^1/2)

Waist and height are in meters and BMI in kg/m², so typical values are small numbers around 0.08 (Krakauer 2012). In the study sample the average ABSI was 0.0808, with a standard deviation of 0.0053 (Krakauer 2012). We convert inches and pounds to meters and kilograms first.

For example, a person 175 cm tall weighing 80 kg (BMI 26.1) with a 100 cm waist has ABSI = 1.00 ÷ (26.1^2/3 × 1.75^1/2) = 1.00 ÷ (8.80 × 1.32), about 0.0859. The paper prints no worked example; this is worked from the formula.

In the study, ABSI had almost no link with height, weight or BMI, while waist size on its own was closely tied to BMI (Krakauer 2012). That is the point of ABSI: two people with the same BMI can have different body shapes.

Your z score: compared with people your age and sex

Average ABSI rises from midlife into old age and is higher in men than in women after young adulthood (Krakauer 2012). So the study did not use raw ABSI to compare people. It turned each value into a z score (Krakauer 2012):

z = (ABSI − average ABSI for your age and sex) ÷ standard deviation for your age and sex

The calculator uses the smoothed averages and standard deviations the authors published for each year of age and each sex, from NHANES 1999 to 2004, in the paper's Table S1 (Krakauer 2012). The table's last row, age 85, covers everyone 85 and over (Krakauer 2012). Some values from it:

Units are m^11/6 kg^−2/3. From Table S1 of the paper, smoothed columns (Krakauer 2012).

What the 2012 study found

Over about 5 years, the risk of death from any cause rose by 33% for each one standard deviation higher ABSI z score (95% CI 20% to 48%), while BMI and waist z scores on their own showed no significant straight-line link (Krakauer 2012). The link with ABSI held after allowing for sex, ethnicity, smoking, diabetes, blood pressure and cholesterol (Krakauer 2012). The authors split people into fifths by ABSI z score and compared each with the middle fifth (Krakauer 2012):

From Table 4 of the paper, which gives the cut points between fifths (Krakauer 2012). The lower two fifths were not significantly different from the middle; the top two had a higher death rate (Krakauer 2012). The authors estimated that 22% (8% to 41%) of the death risk in the population was attributable to high ABSI, compared with 15% for high BMI and 15% for high waist size (Krakauer 2012). The calculator shows which fifth your z score falls in; that is a comparison with this group, not your own risk.

A research marker, with limits

The study followed people for only about 5 years, and ABSI predicted deaths in white and Black adults but not in Mexican American adults (Krakauer 2012). The authors called for more cohorts to define where ABSI is useful (Krakauer 2012). Because ABSI varies over a narrow range, it is sensitive to how accurately the waist is measured; they advise measuring the waist the NHANES way, just above the top of the hip bone, to compare with their reference values (Krakauer 2012). The study cannot show that lowering ABSI lowers risk; that would need trials (Krakauer 2012). No guideline body sets an ABSI cutoff. A high value is a reason to talk to your doctor about your waist size and other heart and metabolic risk factors, not a diagnosis.

Related checks

The BMI calculator measures body size, which ABSI is built to set aside. The waist-to-height ratio has guideline bands from NICE, and the waist-to-hip ratio compares waist with hips. The body roundness index is another body shape score, from waist and height only, and the relative fat mass calculator uses the same two measurements to estimate body fat percentage.