The medical class and the community of statesmen have highlighted the many limitations of the BMI method.
The mathematician Keith Devlin and the association "Center for Consumer Freedom" claim that the margin of error of the BMI is extremely significant, to the point of not even being useful for the evaluation of the state of health.
Political science professor Eric Oliver, of the University of Chicago, argues that BMI is a comfortable but imprecise measure, limited to the population, and therefore should be reviewed.
BMI defects in relation to mathematics and physical characteristics
Since the BMI depends on the weight and the square of the stature but ignores the basic laws of scale referring to the linear dimensions, the highest individuals, even with the same proportions and density compared to lower ones, always have a higher BMI.
The BMI does not take into account body size; a person may have a slender constitution and a morphological, slender type and have more fat than normal despite having a BMI of between 18.5 and 24.9. On the contrary, a robust individual with a short-string morphological type may be in good health, with a fairly low percentage of body fat, but may be classified as overweight due to a BMI equal to or greater than 25. It therefore requires the additional measurement of some body characteristics in order to frame the constitution and the morphological type.
The BMI does not take into account the loss of height with aging. In this circumstance, the BMI increases without any weight gain.
The denominator of the BMI is questionable
The exponent of 2 in the denominator of the BMI formula is arbitrary. It is intended to reduce the variability in the BMI associated only with a difference in objective dimensions, rather than a weight difference relative to one's desirable physiological value. The appropriate exponent should be 3, so the weight would increase with the height cube. However, on average, taller people have less heavy constitution and morphological type than lower people; therefore, the best exponent that would correspond to this variation should be less than 3. An analysis based on data collected in the USA suggested an exponent of 2.6 for subjects from 2 to 19 years, while for adults the exponent could be of 1.92-1.96 for males and 1.45-1.95 for females. The exponent 2 is used only for convention and simplicity.
BMI does not differentiate between lean and fat mass
The hypotheses on the distribution between muscle mass and BMI fat mass are inaccurate. Generally, the BMI overestimates the adiposity of people with more lean body mass (for example, athletes) and underestimates excess fat for those with less lean mass.
A study of June 2008, conducted by Romero-Corral et al., Examined 13, 601 subjects finding (by means of BMI) obesity in 21% of men and in 31% of women. Using body fat percentages (BF%) instead, it was found that obesity affected 50% of men and 62% of women; the correlation between the two evaluations found a greater response in individuals with a BMI between 20 and 30. For men with a BMI of 25, about 20% had a body fat percentage of less than 20% and about 10% had a body fat percentage of over 30%.
BMI is particularly inaccurate to subjects who are in good muscular shape, who may be overweight even with percentages of fat that fall within 10-15%. The body composition of the athletes is best calculated using the estimate of body fat, such as skin folds or hydrostatic weighing or bioimpedance analysis. However, recent studies conducted on American football players who undergo intensive muscle training have shown that they often present the same complications as overweight subjects (eg sleep apnea).
Variation in the definition of BMI categories
It is not yet clear where to place the threshold of overweight and obesity with respect to the BMI scale; for this reason, numerous variations have been applied over the last few decades. Between 1980 and 2000, the "US Dietary Guidelines" framed the overweight within a range between 24.9 and 27.1. In 1985, the "National Institutes of Health" (NIH) defined that overweight should be set by a minimum BMI of 27.8 for men and 27.3 for women. In 1990, the "World Health Organization" (WHO) decided that a BMI between 25 and 30 should be considered overweight and a BMI over 30 as an obese. This has become the definitive guide to determining the presence or absence of overweight. Current WHO and NIH normality scales are associated with reducing the incidence of some diseases such as type II diabetes; however, using the same BMI scale for both men and women is considered a questionable system.
Changes in health status
A study published in 2005 by the "Journal of the American Medical Association" (JAMA) showed that overweight people have a mortality risk very similar to that of normal people (according to the BMI), while underweight and obese people possess a higher mortality rate.
Furthermore, a high BMI is associated with the onset of type 2 diabetes ONLY in people with high serum levels of gamma-glutamyl transpeptidase.
In an analysis of 40 studies involving 250, 000 people, patients suffering from coronary artery disease and having normal BMI were at a higher risk of death (for the same cause) than those with a BMI equal to overweight (BMI 25-29 9).
In a study carried out on persons falling within the BMI range 25-29.9, it was shown that this system was not able to discriminate the percentage of body fat and lean mass. The study concluded that the accuracy of BMI in the diagnosis of obesity is limited, particularly for individuals with intermediate BMI, male and elderly. These results can help explain the reason for the greater statistical longevity of overweight subjects.
A 2010 study that observed 11, 000 subjects for eight years concluded that BMI is not a good assessment system for heart attack, stroke or death risk; on the contrary, the relationship between waist circumference and stature could be. A 2011 study that observed 60, 000 people for 13 years found that the relationship between waist circumference and hip circumference is a better indicator of mortality from ischemic heart disease.
As a possible alternative to BMI, in 1990 the concepts of: Free Fat Mass Index (FFMI) and Fat Mass Index (FMI) were proposed; furthermore, in 2012, the Body Shape Index was proposed.
