Measuring Waist Circumference: Where the Tape Goes According to WHO and NIH Protocols

Where you put the tape measure changes everything. Around the narrowest part of your waist? Over your belly button? At the level of your hip bones? The answer depends on which method you follow, and the difference can shift your result by several centimetres. Two authoritative protocols exist: the WHO STEPS method and the NIH/NHLBI method. Each uses a different anatomical site. Each produces a different number on the same abdomen. Know which one you are using before you compare your result against any published cut-off.

No single site is universal. The World Health Organization (WHO) and the US National Institutes of Health (NIH) / National Heart, Lung, and Blood Institute (NHLBI) specify different anatomical sites, and those sites yield different results on the same abdomen.

WHO STEPS method (also used by NICE CG189): the midpoint between the lower margin of the last palpable rib and the top of the iliac crest. Find the bottom of your ribcage at the side of your torso, then find the top of your hip bone (iliac crest). Mark the halfway point. Place the tape horizontally at that level. The WHO STEPS manual calls this the mid-axillary line, the line running vertically down the side of your body from your armpit.

NIH/NHLBI method: the horizontal plane at the level of the iliac crest. Locate the top of your hip bone and wrap the tape around at exactly that height. This is the site used in the US Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults (1998).

These two sites are not the same spot. The WHO midpoint sits higher on most torsos than the iliac crest. Wang et al. (2003) found that the variation between the narrowest waist and the iliac crest produced a 2-6 cm difference in women and a 1-3 cm difference in men. The WHO site gives a larger figure than the NIH site. The same person can fall above or below a risk threshold depending solely on which method they used.

Locate The Two Bony Points

1. Stand with feet together and arms at your sides. Wear no clothing or thin, non-compressive clothing that does not alter your natural shape.
2. Find the lower margin of your last palpable rib. Press gently at the side of your torso until you feel the bottom edge of your ribcage.
3. Find the top of your iliac crest. Press at the top of your hip bone until you feel the bony prominence.
4. Mark the midpoint between these two points. A measuring tape laid vertically along the mid-axillary line helps you see the halfway point.

Wrap, Breathe, And Read

1. Wrap a flexible, non-stretch tape measure around your abdomen at that midpoint level. Keep the tape horizontal (parallel to the floor) and snug against the skin. It must stay in place without compressing the skin. A Gulick tape measure, which has a tensioning device, standardises this step in clinical settings.
2. Take the reading at the end of a normal exhalation (end-expiration). Do not hold your breath or suck in your stomach. Read to the nearest 0.1 cm.
3. Repeat twice. If the two readings differ by more than 1 cm, take a third and average the two closest.

The respiration phase matters. Your abdomen expands and contracts with each breath. Measuring at end-expiration gives a consistent baseline that minimises the effect of abdominal distension from air or food. Inter-observer error for this method in untrained settings falls between 1.5 and 4.0 cm, according to Verweij et al. (2013). Intra-observer error (the same person measuring themselves twice) is smaller, in the range of 0.5 to 2.0 cm.

Find The Crest And Place The Tape

1. Stand with feet together and arms at your sides. Wear minimal clothing.
2. Locate the top of the iliac crest. This is the bony ridge at the top of your hip bone, easiest to feel when you press in from the side.
3. Wrap a flexible, non-stretch tape measure around your abdomen horizontally at the level of the iliac crest. The tape must be parallel to the floor at all points. It is common for the tape to tilt downward across the back, which changes the reading.
4. Apply consistent tape tension: snug but not compressing the skin. A Gulick tape measure eliminates guesswork here.
5. Take the reading at the end of a normal exhalation. Read to the nearest 0.1 cm.
6. Repeat twice and average the readings.

The NIH/NHLBI site is simpler to locate than the WHO midpoint, which partly explains why it was chosen for the US clinical guidelines. Simplicity comes with a trade-off: it measures a lower point on the torso. In many people that point includes more of the hips and gluteal region, not purely abdominal circumference. The difference between the two sites becomes clinically meaningful when you compare your result against the risk thresholds.

Issuing Body	Anatomical Site	Men (Increased Risk)	Men (High Risk)	Women (Increased Risk)	Women (High Risk)
WHO (2000)	Midpoint lower rib to iliac crest	≥94 cm	≥102 cm	≥80 cm	≥88 cm
NIH/NHLBI (1998)	Iliac crest	Not defined separately	>102 cm	Not defined separately	>88 cm
WHO Asian Populations (2004)	Midpoint lower rib to iliac crest	None defined	≥90 cm	None defined	≥80 cm

The WHO STEPS waist measurement method is part of the Stepwise Approach to Surveillance, a standardised system for collecting noncommunicable disease risk factors across countries. Its site identification procedure is explicit: the midpoint between the lower margin of the last palpable rib and the top of the iliac crest, measured along the mid-axillary line. The method also specifies that the tape measure should be placed directly on the skin if possible, or over thin clothing if privacy requires it, and that the reading should be taken twice to the nearest 0.1 cm.

The WHO method is the basis for the NICE CG189 (2022) guideline, which introduced waist-to-height ratio as a central adiposity marker alongside BMI. NICE states that waist-to-height ratio should be kept below 0.5 as a population-level health signal. The evidence behind that cut-off comes from the Ashwell et al. (2012) systematic review and meta-analysis of 31 studies covering approximately 300,000 adults, which found that waist-to-height ratio is a better screening tool than waist circumference or BMI alone for cardiometabolic risk factors.

How The Site Choice Affects Risk Classification

Measure at the iliac crest (NIH method) and get 90 cm as a man. The WHO increased-risk cut-off is 94 cm at the midpoint site. That 4 cm gap can mean the difference between being told you are at increased risk and being told you are not. Conversely, measure at the WHO midpoint and get 100 cm as a man: you are below the WHO substantially increased risk cut-off of 102 cm. Measure the same person at the iliac crest and the figure is 98 cm. You are now below the NHLBI high-risk cut-off of >102 cm. The interaction of site and threshold is confusing. That is exactly why you need to know which method you followed before you compare your number against any published cut-off.

The NIH/NHLBI waist circumference cut-offs (≥102 cm for men, ≥88 cm for women) were derived from white populations in the 1998 Clinical Guidelines. The WHO cut-offs (≥94 cm and ≥102 cm for men, ≥80 cm and ≥88 cm for women) come from the 2000 Technical Report Series and are intended for global use, though they were also heavily influenced by European data. For Asian populations, the WHO Expert Consultation (2004) recommended lower cut-offs: ≥90 cm for men and ≥80 cm for women, based on the observation that cardiometabolic risk increases at lower waist circumferences in Asian groups.

The NIH/NHLBI chose the iliac crest as the waist circumference site for a practical reason: it is the easiest bony point to locate reliably, even for untrained measurers. The 1998 Clinical Guidelines state that the tape should be placed in a horizontal plane at the level of the iliac crest, and that the reading should be taken at the end of a normal expiration. No mention is made of the midpoint between rib and iliac crest. The US method is simpler and faster, which suits population screening where thousands of readings are taken by staff with minimal training.

But the NIH site is not the same as the WHO site. This mismatch creates confusion in clinical practice, research, and self-measurement. A person who learns the WHO method from a UK or European source and then compares their number against a US cut-off will get a misleading result. The NIH/NHLBI waist circumference cut-off for women is >88 cm (high risk). The WHO cut-off for women is ≥88 cm (substantially increased risk). The difference is a single centimetre and a single word, but it changes whether you are classified as high risk or not.

Reliability And Tape Tension

Inter-rater reliability is a known problem for waist circumference regardless of site. Verweij et al. (2013) reported inter-observer errors between 1.5 and 4.0 cm, with the largest errors occurring when the site was poorly identified. Tape tension was another source of variation: too loose and the reading reads high; too tight and it reads low. The Gulick tape measure, which incorporates a spring-loaded mechanism to apply a standard tension of 100 grams, is the recommended tool in research settings because it eliminates the tension variable. In practice, most people use a standard sewing tape measure and guess at tension, which adds error.

Waist circumference error is not a theoretical possibility. It is the norm when the tape is placed by someone who has not been trained. The most frequent errors:

• Wrong site: placing the tape at the narrowest visual waist, at the belly button (umbilicus), or at the top of the hip rather than at the specified anatomical point. The narrowest waist is what most people think of as their waistline, but it is not what the WHO or NIH methods require. Using the narrowest point instead of the iliac crest or midpoint changes the reading by 2-6 cm in women and 1-3 cm in men, according to Wang et al. (2003).
• Inconsistent respiration phase: measuring after a deep breath in, after a forced breath out, or while holding the breath. The abdomen expands during inhalation and contracts during exhalation. Measuring at the wrong phase adds variability that can exceed 2 cm.
• Incorrect tape tension: pulling the tape too tight (compressing the skin) or leaving it too loose (allowing it to sag). The Gulick tape measure solves this, but it is rarely used outside research.
• Non-horizontal tape: letting the tape dip in the front or rise in the back. The tape must be parallel to the floor at every point around the torso.

Self-measured waist circumference is particularly unreliable. The error commonly exceeds 3-5 cm, and the direction of the error depends on the user's misunderstanding of the site. A person who measures at the narrowest waist will get a smaller number than someone who measures at the iliac crest. A person who measures at the umbilicus may get a larger number. The systematic review by Connor Gorber et al. (2007) found that self-reported height and weight systematically bias BMI downward by 0.5-1.5 kg/m². The same principle applies to waist circumference, where the error is larger because the method is harder to follow. Self-measurement is not equivalent to a protocol-adherent reading, and it should not be treated as such in clinical or research settings.

The site choice does not just produce a different number. It produces a different risk classification. The evidence base for waist circumference as a predictor of cardiometabolic risk is built on specific measurement methods. If you use a different site, you are not measuring the same thing the evidence was built on.

Pischon et al. (2008), in the EPIC cohort of 359,387 adults from 9 European countries, found that each 5 cm increase in waist circumference was associated with a relative risk of 1.17 for men and 1.13 for women for all-cause mortality, adjusted for BMI. The INTERHEART study (Yusuf et al., 2005), covering 27,098 participants from 52 countries, found that waist-to-hip ratio had a graded association with myocardial infarction: the top quintile versus the bottom quintile had an odds ratio of 2.52, adjusted for other risk factors. Both studies used standardised measurement methods, but the site used in each is not always reported in the same way. That makes cross-study comparison difficult.

The NICE CG189 waist-to-height ratio cut-off of 0.5 uses whichever method is stated by the issuing body, usually the WHO midpoint. If you measure at the iliac crest and apply the 0.5 rule, you are using a cut-off that was validated on a different measurement. It may or may not give you the same answer. The safest approach: state the method you used every time you report a waist circumference number, so that the reader (or the clinician) knows which cut-off to apply.

Site identification is the largest single source of error in waist circumference measurement. The person who places the tape at the narrowest visual waist, at the umbilicus, or at the top of the hip rather than at the specified site is not following either method. The number they produce is not comparable to the published cut-offs. The inter-observer error of 1.5-4.0 cm reported by Verweij et al. (2013) assumes the measurer knows which site to use. If the site is misunderstood, the error can be much larger.

Self-measured waist circumference error commonly exceeds 3-5 cm. The direction of the error depends on the user's misunderstanding of the site. A person who measures at the narrowest waist will get a smaller number than someone who measures at the iliac crest. A person who measures at the umbilicus may get a larger number. The single thing that most often goes wrong: the tape is placed at the wrong site, and the person interpreting the number does not know which site was used.