Measuring Adiposity in Patients: The Utility of Body Mass Index (BMI) & Body Fat | Hacker News Reader

Measuring Adiposity in Patients: The Utility of Body Mass Index (BMI) & Body Fat

journals.plos.org

36 pointsgraeme30 comments

as this study shows, BMI is a useful measure despite its limitations, however others have preferred waist-to-hip ratio as a better health risk metric.

waist-to-hip-ratio is actually simpler to measure and calculate than BMI.

Here’s how from WebMD: https://www.webmd.com/fitness-exercise/what-is-waist-to-hip-...

Measure waist at the smallest point. Wrap a tape measure around your waist at the smallest point, usually around your belly button. This is your waist circumference.

Measure hips by wrapping the tape measure around your hips at the widest part. This is your hip circumference.

Divide your waist size by your hip size. This is your waist-to-hip ratio. A good rule of thumb is that your waist should be smaller than your hips, no matter your weight or BMI.

WHO defines abdominal obesity in men as a waist-to-hip ratio of at least 0.90. For women, it’s a ratio of 0.85 or more. A ratio higher than 1.0 for either sex means a much higher chance of health problems.

TFA mentions waist-to-hip but doesn't assess it: "We were not able to accurately capture co-morbidities. We were also unable to compare other anthropometric indices, such as waist-to-hip ratio with corresponding DXA measurements, due to lack of hip circumference data."

more from WebMD:

Too much fat around your waist is known as abdominal obesity. This fat is called visceral fat because it surrounds your liver and other organs. This can affect both men and women, but men are more likely to have too much fat around the waist. Women are more likely to carry fat around their hips and thighs.

Visceral fat sends hormones, fatty acids, and other chemicals that cause inflammation into your body. This leads to higher cholesterol, blood pressure, and blood glucose. It also leads to higher levels of triglycerides, a type of fat, in your blood.

Fat around your waist can make you more likely to have other health problems, including: Heart disease, Cancer, Stroke, Type 2 diabetes

I simply cannot imagine such a measure to be sensible for all types of body shapes.

I linked to one review earlier, but there's actually a lot of evidence that WHtR is quite good against a number of populations and ages, and certainly for each individual, even if you ignore cutoffs, it will be directionally correct:

Corbatón Anchuelo, Arturo, María Teresa Martínez-Larrad, Irene Serrano-García, Cristina Fernández Pérez, and Manuel Serrano-Ríos. “Body Fat Anthropometric Indexes: Which of Those Identify Better High Cardiovascular Risk Subjects? A Comparative Study in Spanish Population.” PLoS ONE 14, no. 5 (May 23, 2019). https://doi.org/10.1371/journal.pone.0216877.

Djap, H. S., B. Sutrisna, P. Soewondo, R. Djuwita, K. H. Timotius, Trihono, S. Sharif, and Y. S. Tjang. “Waist to Height Ratio (0.5) as a Predictor for Prediabetes and Type 2 Diabetes in Indonesia.” IOP Conference Series: Materials Science and Engineering 434 (December 2018): 012311. https://doi.org/10.1088/1757-899X/434/1/012311.

Zhu, Qihan, Feixia Shen, Tingting Ye, Qi Zhou, Huihui Deng, and Xuejiang Gu. “Waist-to-Height Ratio Is an Appropriate Index for Identifying Cardiometabolic Risk in Chinese Individuals with Normal Body Mass Index and Waist Circumference.” Journal of Diabetes 6, no. 6 (November 2014): 527–34. https://doi.org/10.1111/1753-0407.12157.

Vrdoljak, Davorka, Biserka Bergman Marković, Ksenija Kranjčević, Dragica Ivezić Lalić, Jasna Vučak, and Milica Katić. “How Well Do Anthropometric Indices Correlate with Cardiovascular Risk Factors? A Cross-Sectional Study in Croatia.” Medical Science Monitor : International Medical Journal of Experimental and Clinical Research 18, no. 2 (February 1, 2012): PH6–11. https://doi.org/10.12659/MSM.882451.

Ochoa Sangrador, C., and J. Ochoa-Brezmes. “Waist-to-Height Ratio as a Risk Marker for Metabolic Syndrome in Childhood. A Meta-Analysis.” Pediatric Obesity 13, no. 7 (2018): 421–32. https://doi.org/10.1111/ijpo.12285.

Caminiti, Carolina, Marisa Armeno, and Carmen S. Mazza. “Waist-to-Height Ratio as a Marker of Low-Grade Inflammation in Obese Children and Adolescents.” Journal of Pediatric Endocrinology & Metabolism: JPEM 29, no. 5 (May 1, 2016): 543–51. https://doi.org/10.1515/jpem-2014-0526.

Thank you all for your replies.

The claim was just that it’s a better indicator than BMI, which is certainly not sensible for all bodies.

I think the other replies to you are good, i'll just add,

unlike BMI, waist to hip focuses on abdominal fat and the additional claim is that abdominal fat is "the worst kind" for health, i.e. (i'm not an expert, and i'm not going to research, but from memory) men tend to get belly fat, women tend to get hip fat, but women who get belly fat are at increased health risk in the same way that men are, heart attacks, etc.

DXA scans are available (in the US at least) from people like BodySpec. When I started a weight loss program I found the DXA scan to be the most helpful data in terms of understanding how things were changing given my changes in diet and exercise regimen. I tend to be somewhat data driven :-) and that definitely scratched the data itch. It also gave me a better target (body fat % / adipose tissue target) to go for rather than a specific body weight. In the past it has always seemed arbitrary to me that someone "should" weigh, X pounds given their height.

I used BMI as a tool to see how much I was losing since I started eating less, to help me set goals and reach 'healthy' levels.

32->28 took a bit more than a year. I remember I weighted myself because I was able to walk two hours without ankle pain (the last hour was a bitch), and I started doing sports other than sailing/windsurfing/kayaking.

The 28->26 took three years, with a lot more sport than I used too. I'm still overweight but my waist/hip ratio is under 1 (barely) and I have less ankle pain and respiratory issues than I used too. I think I really stopped being obese in this range.

The problem isn't so much the BMI, as the idea that obesity starts at BMI 30.

Probably, say, 22-25 should be overweight, 25+ obese.

22 at average western height is a perfectly healthy weight.

A typical male weighing 70kg at a height of 1.80m has basically no excessive fat and a BMI of 21.6.

What BMI needs is a more refined scale for different kinds of humans.

Sorry, to clarify; the overweight tipping point might be found somewhere in the range 22-25, not that this is to be the overweight range. If this hypothetical person puts on 10 kg of fat, that almost brings him to 25. Say 5 kg is overweight, 10 kg the start of obesity. Looks reasonable.

Seems kind of unreasonable that one can go from "normal" to "obese" by putting on 10kg. Feels like that's just removing the entire overweight range.

The idea that 25 should be where obesity starts is frankly absurd.

Why should there be a large overweight range of more than some 6-8 kg (for that 1.8 m height)? That just sounds like fat acceptance.

10 kg is like a toddler.

Why should there be a large obese range (80kg to infinity), but a tiny overweight range of 5kg? What is even the point of the overweight range in that context?

It's not fat acceptance. Being overweight is still unhealthy.

When a parameter switches between radically different interpretations along its value range, it's sometimes useful to have a reasonable transition area to eliminate the abrupt change.

You know, like a yellow light between green and red.

For you maybe. BMI is a terrible generalistic measure to make specific recommendations.

There is strong relationship between increased leptin and increased percent body fat, and no relationship between insulin and percent body fat. Error bars represent 95% confidence intervals for mean.

And

BMI significantly underestimates prevalence of obesity when compared to DXA direct measurement of percent body fat. Currently, no other blood test or biomarker has been correlated with the rate of obesity. The use of both DXA and leptin levels offers the opportunity for more precise characterization of adiposity and better management of obesity.

Finally

The use of leptin levels further improves precision of BMI adjustment, whereas insulin levels do not. With 91% of our patients with high leptin levels being women, our data confirm the greater effectiveness of BMI adjustment with leptin levels in women, attributable to a higher prevalence of hyperleptinemia among women. As significant lowering of leptin impacts long term weight control [31], [32], the idea of incorporating leptin adjustments into a more accurate diagnosis of obesity should be seriously considered. Further studies should be conducted for leptin measurements as a potentially useful tool in the management of obesity.

> BMI characterized 26% of the subjects as obese, while DXA indicated that 64% of them were obese. 39% of the subjects were classified as non-obese by BMI, but were found to be obese by DXA. BMI misclassified 25% men and 48% women.

Just goes to show how unreliable BMI is… oh wait

“Happily obese” people like to point out how wrong BMI is. Little did they know they were right for the wrong reason.

Is there anybody (obese and non obese) in the health community that assumes BMI isn‘t wrong ?

I’m not sure what your point is.

I’m surprised this needs to be explained, but the joke is that every time BMI comes up in online discussions, people rally to annonuce that it’s too conservative because it classifies some dedicated bodybuilders as obese, while this study shows that it’s aggregate bias seems to be in precisely the opposite direction.

Which, of course, better fits everyday intuition about how unhealthy most people with high BMI seem to be.

In reality, the failure of BMI to accurately capture the body fat and healthfulness of most individuals is not usually because they’re just too damned ripped. It’s because people eat more and sit more than they ever have.

Yep just below this thread there's a [dead] comment about how a bodybuilder had high BMI despite not being fat. It's always the same cliche.

Pointing out that Earth is round no matter how often it is repeated does not become any less true.

BMI and bodybuilder example may be less obvious unless you know that muscles are more dense than fat but it is true nonetheless.

> people rally to annonuce that it’s too conservative

That’s a bit of a surprise to me. The most common conplaint I heard was “it’s wrong, but we need to keep it around otherwise there’s no way to compare cross-studies”.

We have decades and decades of BMI based classification, and publishing a paper today that needs results from past studies will automatically bind it to BMI. Same on public policies, defining obesity threshold a completely different way would require more politics than most groups are able to tackle.

It’s like coming up today with a different way to count nutrients from calories, technical merit won’t make it prevail over the whole world that has been built around calories.

The paper mentions that "In 1995, the World Health Organization (WHO) defined obesity based on a percent body fat ≥25% for men and ≥35% for women [15], while the most recent 2009 guideline [..] used percent body fat ≥25% for men and ≥30% for women". I wonder what the process was for choosing the BMI and obesity cut-points, since they seem to differ a lot.

BMI has been known to be unreliable for almost as soon as it was published, but people clinged to it because they could get some number with absolutely no education and only basic tools on the measuring side. You could do it in the jungle with a tape measure and a weighting board.

So the base reasonning was “a somewhat dummy number is better than no number at all”.

DXA is 100% better but requires trained personnel and expensive, advanced scanners. I’m not sure it’s a fitting upgrade of BMI…and if a subject can be thorougly scanned, can’t they also get more personal advice and actual assessment of what they are risk of (blood pressure ? heart disease ? diabetes ?) instead of some vague “you’re fat so you should lose weight, we’ll see how you’ll actually die once you’re thinner” ?

There's a lot of misinformation about BMI... Is it perfect? No, in fact for women, it tends to significantly under-report obesity (when defined against fat percentage, officially, obesity is mostly defined by BMI though so of course it's always accurate in that sense). It's actually quite accurate for men though.

There has in fact been extensive study on better formulas, and one recent study has proposed Relative Fat Mass, RFM as a new index that does a better job correlating to DXA measurements for all populations:

Woolcott, Orison O., and Richard N. Bergman. “Relative Fat Mass (RFM) as a New Estimator of Whole-Body Fat Percentage ─ A Cross-Sectional Study in American Adult Individuals.” Scientific Reports 8, no. 1 (July 20, 2018): 10980. https://doi.org/10.1038/s41598-018-29362-1.

Someone else mentioned waist-to-height-ratio (WHtR) as a better marker for metabolic health, and I tend to agree that it's superior to BMI:

Ashwell, Margaret, and Sigrid Gibson. “Waist-to-Height Ratio as an Indicator of ‘Early Health Risk’: Simpler and More Predictive than Using a ‘Matrix’ Based on BMI and Waist Circumference.” BMJ Open 6, no. 3 (March 14, 2016). https://doi.org/10.1136/bmjopen-2015-010159.

If you're also arguing that people should be looking at other markers for metabolic health, sure. If you get an annual checkup, you should get everything you need to calculate your metabolic syndrome markers, which is an excellent way to assess metabolic health:

Huang, Paul L. “A Comprehensive Definition for Metabolic Syndrome.” Disease Models & Mechanisms 2, no. 5–6 (2009): 231–37. https://doi.org/10.1242/dmm.001180.

> It's actually quite accurate for men though.

https://pubmed.ncbi.nlm.nih.gov/15807202/

> In this study BMI shows a low reliability as a predictor of individual body fatness, particularly in men and in subjects with a BMI below 30 kg/m2.

> waist-to-height-ratio (WHtR)

Wouldn't elecricatl impedance analysis yield better results, as we have most of our home weighting board at this point ? I mean, for now a pretty low price we have better device to get approximation than a measuring tape.

> No

There's no point to having further factual discussion if you aren't going to read the referenced papers. From the 2018 study I referenced you can see that DXA-positive and BMI-positive correlation was 2,668/2,778 (Table 3), or simply look at Figure 2.

And if you say "No" and then link to a 2004 paper using bio-electrical impedance, I don't really know what to say, since BIA is one of the least accurate methods and there's plenty of research comparing indices to gold-standard DXA already.

Nickerson, Brett S. “Agreement between Single-Frequency Bioimpedance Analysis and Dual Energy x-Ray Absorptiometry Varies Based on Sex and Segmental Mass.” Nutrition Research (New York, N.Y.) 54 (2018): 33–39. https://doi.org/10.1016/j.nutres.2018.03.003.

In fact, here's a paper concluding recommending BIA shouldn't be used clinically from ... 2004:

Buchholz, Andrea C., Cynthia Bartok, and Dale A. Schoeller. “The Validity of Bioelectrical Impedance Models in Clinical Populations.” Nutrition in Clinical Practice: Official Publication of the American Society for Parenteral and Enteral Nutrition 19, no. 5 (October 2004): 433–46. https://doi.org/10.1177/0115426504019005433.

If you've used a BIA measuring scale at home, you'll see that it fluctuates wildly and serves more of a measure of how dehydrated you are during the day than any other health metric.

Ultimately though, I think you're splitting hairs/missing the point. There are tons of bio-markers anyone case use to track their health progress directionally at home. Just pick one and use what you want and stop worrying about what other people are doing. Most people who argue the most against BMI or whatever as a marker already know that they're not as healthy as they should be. And if you don't want to measure anything, if you're an American adult, you can probably just safely assume that you're not (only 12% "metabolically optimal" from 2009-2016 NHANES data):

Araújo, Joana, Jianwen Cai, and June Stevens. “Prevalence of Optimal Metabolic Health in American Adults: National Health and Nutrition Examination Survey 2009–2016.” Metabolic Syndrome and Related Disorders 17, no. 1 (November 27, 2018): 46–52. https://doi.org/10.1089/met.2018.0105.

You made 3 points:

First one:

> There's a lot of misinformation about BMI[...]It's actually quite accurate for men though.

I replied with a detailed study from 2004 on how, no, it’ not quite accurate, even for men. (I didn’t push on it, but how garbage is a population wide indicator that would be accurate mostly for men?)

Second one:

> RFM as a new index that does a better job correlating to DXA measurements for all populations

> Someone else mentioned waist-to-height-ratio (WHtR) as a better marker for metabolic health, and I tend to agree that it's superior to BMI

Both are indexes based on external size measurements, from a body of subjects that has 70+% european american, 70% of the total with a BMI bellow 30. From CDC’s numbers, 40+% of the US population is above 30 of BMI (https://www.cdc.gov/obesity/data/adult.html) so the conclusions we can draw from these results are pretty limited (the lack of diversity in the panel is further mentioned as a limitation)

WHtR is a different philosophy altogether and is used as a trigger value and not an indicator.

> If you get an annual checkup, you should get everything you need to calculate your metabolic syndrome markers, which is an excellent way to assess metabolic health

In an annual checkup I’m not sure how relevant metabolic assessment is. My point was indicators are useful to decide on further and potentially more costly/invasive analyses. If you’re already getting a full checkup with blood analysis, echo scans and all, a metabolic score won’t be the primary trigger for more action.

> BIA

I’m not saying it should be used in lieu of more accurate scans, but I see them as more reliable than RFM for instance. Variability throughout the day and weakness to water levels will also affect the other external measurement methods.

> Most people who argue the most against BMI or whatever as a marker already know that they're not as healthy as they should be

BMI reliance has real world impacts. Every year people will be sent to extra checkups, health programs etc. because of the BMI cutouts. It’s not some cute number people ignore or not, time and money and potentially extra things like company programs and insurance premiums will be affected depending on the bucket you fall in.

Reply on news.ycombinator.com