Body Mass Index / Quetelet Index

1. Abstract

The Body Mass Index (BMI), originally conceptualized as the Quetelet Index by the Belgian mathematician, astronomer, and statistician Lambert Adolphe Jacques Quetelet in the 1830s and formalized in 1842, is an anthropometric measure defining the ratio of an individual’s body weight to the square of their height (expressed as $\text{kg/m}^2$). Although initially designed within the paradigm of “social physics” to delineate the physical characteristics of the average human (l’homme moyen), the index was repositioned as a clinical and epidemiological biomarker of relative adiposity by Ancel Keys and colleagues in their landmark 1972 investigation. In modern clinical psychology, behavioral medicine, and psychiatric diagnostics, the BMI serves as an indispensable quantitative tool for operationalizing nutritional states, diagnosing and staging eating disorders under the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR), and evaluating the somatic correlates of severe psychopathology, psychotropic medication-induced metabolic shifts, and body image disturbance.

As an anthropometric instrument, BMI does not rely on self-report item banks or Likert rating scales; rather, it is a continuous physical metric categorized into standardized operational thresholds defined by the World Health Organization: underweight ($<18.5\text{ kg/m}^2$), normal weight ($18.5\text{–}24.9\text{ kg/m}^2$), pre-obesity/overweight ($25.0\text{–}29.9\text{ kg/m}^2$), and obesity classes I, II, and III ($ge 30.0\text{ kg/m}^2$). Psychometrically, when examined as a proxy indicator for body adiposity, BMI demonstrates strong convergent validity with gold-standard densitometric and radiographic technologies, including Dual-Energy X-ray Absorptiometry (DEXA) and hydrostatic weighing, yielding correlation coefficients typically ranging from $r = 0.70$ to $0.90$. Its test-retest reliability across objective measurements is extraordinarily high ($ICC > 0.98$), though self-reported values display systematic reporting biases. This comprehensive review analyzes the historical, theoretical, psychometric, and clinical frameworks governing the Quetelet Index, emphasizing its contemporary utility and limitations across psychological assessment and behavioral healthcare.

2. Keywords

Body Mass Index, Quetelet Index, Adolphe Quetelet, Ancel Keys, Anthropometry, Eating Disorders, Anorexia Nervosa, Body Image Dissatisfaction, Obesity, Behavioral Medicine, Psychometrics, Body Composition

3. Authors

The theoretical, mathematical, and observational foundations of the index were established by:

• Lambert Adolphe Jacques Quetelet (1796–1874): Belgian astronomer, mathematician, statistician, and sociologist. Director of the Royal Observatory of Belgium, Brussels, Belgium. Quetelet introduced the index in his foundational treatise Sur l’homme et le développement de ses facultés, ou Essai de physique sociale (1835) and elaborated upon it in A Treatise on Man and the Development of His Faculties (1842).
• Ancel Keys, Ph.D. (1904–2004) (Modern Clinical Validation): Laboratory of Physiological Hygiene, School of Public Health, University of Minnesota, Minneapolis, MN, USA. Keys and his international collaborators validated and renamed the construct “Body Mass Index” in their 1972 study published in the Journal of Chronic Diseases.

4. Purpose

The primary purpose of the Body Mass Index (BMI) / Quetelet Index across clinical psychology, psychiatric assessment, and behavioral medicine is to provide an objective, rapid, non-invasive, and standardized metric of human mass relative to stature. While originally conceived to mathematically describe cross-sectional population distributions of physical dimensions, contemporary applications leverage BMI as both an independent physical parameter and an explanatory variable within biopsychosocial paradigms. In clinical psychology and psychiatry, the measure is utilized for diagnostic classification, risk stratification, monitoring therapeutic interventions, and examining somatic-behavioral interactions.

Within the domain of eating disorders, BMI represents an essential diagnostic and staging criterion. The American Psychiatric Association utilizes BMI cut-off values in the DSM-5 and DSM-5-TR to operationalize severity levels for Anorexia Nervosa in adults:

• Mild: $\text{BMI} ge 17\text{ kg/m}^2$
• Moderate: $\text{BMI } 16\text{–}16.99\text{ kg/m}^2$
• Severe: $\text{BMI } 15\text{–}15.99\text{ kg/m}^2$
• Extreme: $\text{BMI} < 15\text{ kg/m}^2$

Conversely, in the assessment of Binge Eating Disorder (BED) and non-purging bulimia nervosa, tracking longitudinal BMI trajectories assists clinicians in evaluating comorbid obesity, cardiometabolic risk, and physiological deterioration secondary to dysregulated eating behavior.

Beyond specialized eating disorder clinics, BMI is widely employed in general psychiatric and psychological research. First, it serves as a critical monitoring metric for the adverse metabolic sequelae of psychotropic pharmacotherapy, notably second-generation (atypical) antipsychotics such as olanzapine and clozapine, which are frequently linked to rapid weight gain, insulin resistance, and metabolic syndrome. Second, BMI is an indispensable covariate and moderating variable in empirical studies on weight stigma, internalizing psychopathology, major depressive disorder, self-esteem, neurocognition, and body image dissatisfaction. It enables researchers to isolate psychological variance from physical stature and excess adiposity, clarifying whether emotional distress stems from biological pathology, perceived physical appearance, or institutionalized discrimination.

5. Psychological Construct

At its physical root, the Quetelet Index measures relative gross body mass normalized for vertical skeletal length. However, when integrated into psychological science, behavioral epidemiology, and psychosomatics, the physical metric transforms into a multi-faceted construct intersecting biological reality, cognitive appraisal, and socio-cultural feedback loops. Clinicians and researchers conceptualize BMI as an objective marker that anchors several underlying psychological phenomena:

5.1. Objective Physicality Versus Subjective Body Image

In psychometrics, BMI is often juxtaposed against subjective measures such as the Multidimensional Body-Self Relations Questionnaire (MBSRQ) or the Body Shape Questionnaire (BSQ). This enables the operationalization of body size misperception and body image discrepancy. By calculating the mathematical delta between an individual’s actual BMI and their self-estimated or idealized BMI (often evaluated using computerized silhouette scales or digital body morphing tasks), psychologists quantify perceptual distortion. For instance, in individuals with restrictive anorexia nervosa, high degrees of cognitive distortion and somatic delusion manifest as severe overestimation of personal BMI despite objectively emaciated status.

5.2. Weight Bias Internalization and Social Stigma

BMI operates as an index of visibility within a societal framework that culturally lionizes thinness and stigmatizes adiposity. Consequently, an elevated BMI serves as a significant predictor for experiencing explicit and implicit weight-based discrimination. In psychological testing, BMI is regularly combined with the Weight Bias Internalization Scale (WBIS) to measure how deeply cultural stereotypes are incorporated into an individual’s self-concept. Research confirms that individuals with higher BMI categories who exhibit elevated weight bias internalization report higher incidences of social anxiety, demoralization, experiential avoidance, and elevated cortisol reactivity.

5.3. Allostatic Load and Biological Distress

In health psychology and psychosomatic medicine, extreme deviations in BMI—both severe emaciation ($<16\text{ kg/m}^2$) and morbid obesity ($ge 40\text{ kg/m}^2$)—are conceptualized as biological indicators of chronic distress and neuroendocrine dysregulation. Chronic elevations in BMI correlate with systemic low-grade inflammation (marked by elevated C-reactive protein and interleukin-6), hypercortisolemia, hypothalamic-pituitary-adrenal (HPA) axis dysregulation, and neurostructural changes in prefrontal executive control networks. Consequently, BMI functions not simply as a morphological variable, but as a somatic proxy for allostatic overload, directly influencing mood regulation, impulse control, and stress vulnerability.

6. Theoretical Framework

The theoretical framework underpinning the Body Mass Index draws from historical social physics, physical allometry, and the modern biopsychosocial model of physical and psychological well-being.

6.1. Quetelet’s Allometric Law and “Social Physics”

Lambert Adolphe Jacques Quetelet was motivated by the philosophy that statistical distributions could unveil universal laws governing humanity, mirroring the laws of celestial mechanics. He formulated the concept of l’homme moyen (the average man)—a central tendency representing human physical, intellectual, and moral perfection, around which population deviations occur following the Gaussian normal distribution. During his cross-sectional anthropometric examinations of children, soldiers, and civilian populations in Belgium and France, Quetelet sought to determine the mathematical relationship between longitudinal growth and overall mass accumulation. He observed that while volumetric scaling would theoretically dictate that mass increases with the cube of height ($M propto H^3$), human vertical growth is accompanied by disproportionate transverse development. Specifically, after the completion of adolescent growth, Quetelet observed that body weight increases proportionally not to the cube, but approximately to the square of height:

$$\text{Weight} propto \text{Height}^2 implies \frac{\text{Weight}}{\text{Height}^2} \approx \text{constant}$$

This power-law relationship yielded the Quetelet Index, intended as an allometrically sound method to eliminate the influence of height when evaluating whether an adult’s overall mass conformed to the central tendency of the population.

6.2. Ancel Keys and the Epidemiological Turn

For more than a century, Quetelet’s index remained primarily within the remit of actuarial science and physical anthropology. In 1972, Ancel Keys, along with Noboru Kimura, Henry Blackburn, and Martti Karvonen, conducted an empirical investigation of 7,424 healthy men across five cohorts in four countries (United States, Finland, Italy, and South Africa). Keys sought to resolve a long-standing methodological controversy: which simple anthropometric index of relative weight exhibited the highest correlation with body fatness (measured by body density via underwater weighing and subcutaneous skinfold thickness) while maintaining near-zero correlation with stature?

Keys evaluated several candidates, including the simple weight-for-height ratio ($W/H$), the ponderal index ($H/\sqrt[3]{W}$), and the Quetelet Index ($W/H^2$). The findings conclusively revealed that the Quetelet Index was consistently superior to all other metrics: it demonstrated the lowest correlation with height ($r \approx 0.00\text{ to } -0.09$) and the highest correlation with skinfold thickness and densitometric adiposity ($r \approx 0.70\text{ to } 0.85$). Recognizing its practical utility for population surveillance and clinical screening, Keys officially coined the term Body Mass Index, firmly embedding it into global medical, epidemiological, and psychiatric nomenclature.

6.3. The Biopsychosocial Integration

In contemporary psychological science, BMI is conceptualized through the lens of George Engel’s Biopsychosocial Model. The biological components (genetics, metabolism, neurochemistry) determine the physiological parameters of the index. The psychological components (appetitive traits, self-regulation, eating behaviors, mood states, adverse childhood experiences) drive behavioral patterns that modulate weight trajectories over time. Concurrently, the social components (socioeconomic status, systemic prejudice, food deserts, cultural beauty standards) determine the environmental affordances and interpersonal feedback an individual experiences based on their physical categorization. Therefore, BMI sits at the center of these overlapping domains, functioning as both an input and an output within continuous behavioral feedback loops.

7. Validity

Because the Body Mass Index is an observational anthropometric measure rather than a subjective psychometric scale, its evaluation involves specific criteria: convergent validity with direct body composition assessments, predictive validity for physiological and psychological endpoints, and discriminant validity concerning non-adipose mass components.

7.1. Convergent and Criterion Validity

Extensive biomedical and psychological research has evaluated the convergent validity of BMI against reference standard methods of body composition analysis, including Dual-Energy X-ray Absorptiometry (DEXA), underwater hydrostatic weighing, air displacement plethysmography (Bod Pod), and bioelectrical impedance analysis (BIA).

• In large cross-validation trials, the correlation between BMI and total body fat mass (measured in kilograms via DEXA) is exceptionally high, typically ranging between $r = 0.82$ and $r = 0.94$.
• When correlating BMI with percentage body fat (%BF), coefficients remain robust yet moderate, generally falling between $r = 0.65$ and $r = 0.82$. This slight attenuation occurs because BMI captures total mass, which includes both adipose tissue and fat-free mass (skeletal muscle, bone mineral content, organ weight, and water).
• Studies employing receiver operating characteristic (ROC) analyses indicate that at the standard cutoff of $\text{BMI} ge 30\text{ kg/m}^2$, the index demonstrates high specificity ($>95%$) for identifying excess adiposity (%BF > 25% in men, > 35% in women), but only moderate sensitivity ($50%\text{–}65%$). Consequently, while an individual with a BMI above $30\text{ kg/m}^2$ almost universally exhibits high adiposity, a considerable proportion of individuals with “normal” or “overweight” BMI scores possess elevated body fat percentages coupled with low muscle mass—a clinical presentation frequently termed normal-weight obesity (NWO).

7.2. Predictive and Prognostic Validity

BMI exhibits significant predictive validity across a spectrum of long-term somatic and psychiatric outcomes:

• All-Cause and Cardiovascular Mortality: Prospective epidemiological meta-analyses (encompassing over 10 million participants, such as the Global BMI Mortality Collaboration) document a pronounced J-shaped or U-shaped mortality curve. The lowest hazard ratios for all-cause mortality consistently converge between $22.0\text{ and } 24.9\text{ kg/m}^2$. Significant elevations in mortality risk occur at both extremes ($<18.5\text{ kg/m}^2$ due to malnutrition, sarcopenia, and underlying wasting diseases; and $ge 35.0\text{ kg/m}^2$ due to ischemic heart disease, stroke, and obesity-related malignancies).
• Psychiatric Morbidity: Longitudinal cohort studies indicate that both low BMI in adolescence and severe obesity in adulthood reliably predict the onset of depressive disorders. In eating disorder populations, initial admission BMI, alongside the rate of BMI normalization during refeeding protocols, serves as the most potent prognostic indicator of long-term recovery, relapse likelihood, and bone mineral restoration.

7.3. Discriminant Limitations

The principal critique of BMI’s construct validity centers on its limited discriminant validity regarding somatic tissue distribution. BMI cannot differentiate:

• Hypertrophic lean skeletal muscle mass from pathological adipose tissue (yielding false-positive obesity classifications in bodybuilders and power athletes).
• Visceral adipose tissue (metabolically toxic fat surrounding abdominal organs) from subcutaneous peripheral adipose tissue (relatively benign adipose deposits around the thighs and hips).
• Sex-specific and age-related body composition trajectories, as women naturally carry higher fat mass percentages at equivalent BMI levels compared to men, and aging adults frequently lose muscle mass while gaining visceral adiposity without any observable change in total BMI (sarcopenic obesity).

8. Reliability

In anthropometry and behavioral research, the evaluation of reliability diverges from classical questionnaire psychometrics (such as Cronbach’s $\alpha$ or McDonald’s $\omega$), focusing instead on technical errors of measurement, intra-rater and inter-rater agreement, and the structural reliability of self-reported data versus direct clinical measurements.

8.1. Objective Measurement Reliability

When physical stature and total body weight are acquired directly by trained examiners utilizing standard clinical equipment (e.g., calibrated balance beam scales or Class III digital medical scales, alongside wall-mounted stadiometers):

• Intra-Class Correlation Coefficients (ICC): Test-retest reliability across brief temporal intervals (e.g., within-day or consecutive-day assessments) exceeds $ICC = 0.99$ for both mass and height, yielding a composite BMI test-retest reliability of $ICC > 0.98$.
• Technical Error of Measurement (TEM): In rigorously standardized protocols, the absolute intra-examiner TEM for adult height is typically less than $0.3\text{ cm}$, and for weight it is less than $0.1\text{ kg}$. The resulting relative inter-examiner TEM for calculated BMI consistently falls below $1.5%$, representing exceptional instrument stability.

8.2. Reliability of Self-Reported BMI

In epidemiological surveys, psychiatric registries, and telepsychology studies, direct physical measurement is frequently substituted with self-reported height and weight due to logistical constraints. Psychometric evaluations of self-reported BMI reveal consistent, systematic measurement artifacts:

• Systematic Directional Bias: Across global cohorts, individuals systematically overreport their height (average bias: $+0.5\text{ cm to }+2.5\text{ cm}$) and underreport their body weight (average bias: $-1.0\text{ kg to }-3.5\text{ kg}$). This dual bias produces a downward skew in calculated BMI of approximately $0.8\text{ to } 1.6\text{ kg/m}^2$.
• Differential Validity by Subgroup: The magnitude of self-report underestimation is not random; it correlates positively with true body weight, age, and female gender. Individuals meeting objective criteria for Class II and Class III obesity exhibit significantly larger reporting errors than lean individuals.
• Reliability Attenuation: While the correlation between self-reported and measured BMI remains high ($r = 0.88\text{ to } 0.94$), reliance on self-report systematically misclassifies $15%\text{ to } 25%$ of individuals into lower WHO clinical risk categories, underestimating true obesity rates and attenuating effect sizes in behavioral research.

9. Factor Analysis and Structural Modeling

Because the Quetelet Index is a derived bivariate ratio, it does not possess an internal latent factor structure derived via conventional Exploratory Factor Analysis (EFA) or Confirmatory Factor Analysis (CFA). Instead, structural analysis of the index focuses on allometric power modeling and structural equation modeling (SEM), where BMI serves as an observed continuous variable within broader latent behavioral and psychological architectures.

9.1. Allometric Power Scaling and Structural Orthogonality

The mathematical validity of the formula relies on power-law scaling analysis to ensure that the resulting ratio is structurally orthogonal to height. When analyzing human morphology, the structural equation linking mass ($W$) and height ($H$) is expressed as:

$$W = \alpha \cdot H^p$$

Logarithmic transformation yields the linear structural equation:

$$\ln(W) = \ln(\alpha) + p \cdot \ln(H)$$

In structural evaluations across diverse adult populations, the empirical regression exponent $p$ consistently falls between $1.92$ and $2.15$, confirming that $p = 2$ represents the optimal scaling power to neutralize the correlation between height and weight. If $p=1$ were used ($W/H$), the index would remain positively correlated with height ($r \approx 0.30\text{–}0.45$), biasing classifications in favor of taller individuals. Conversely, if $p=3$ were used ($W/H^3$, the Tri-Ponderal or Rohrer Index), the metric would become negatively correlated with height ($r \approx -0.20\text{ to } -0.35$). Thus, $W/H^2$ exhibits structural optimality by maximizing the adiposity signal while rendering the metric orthogonal to vertical skeletal dimension in adult populations.

9.2. Structural Equation Modeling (SEM) in Psychopathology

In psychometrics and behavioral medicine, BMI is frequently integrated into structural equation models examining eating behavior, depression, and metabolic dysregulation. Standard configurations include:

• The Dual-Pathway Model of Bulimic Pathology: SEM investigations demonstrate that elevated baseline BMI directly influences structural latent factors such as Appearance Ideal Internalization and Body Dissatisfaction (standardized path coefficients typically ranging from $\beta = 0.25\text{ to } 0.40$). Body dissatisfaction subsequently predicts both Dietary Restraint and Negative Affectivity, which directly feed into the latent construct of Bulimic Symptoms.
• Mediation Models of Antipsychotic Weight Gain: Latent growth curve modeling (LGCM) positions longitudinal change in BMI ($\Delta\text{BMI}$) as a central mediator linking genetic polymorphisms (such as 5-HT2C receptor gene variants) to patient-reported medication non-adherence, subjective distress, and subsequent somatic comorbidities.

10. Instrument / Measurement Tool

Unlike psychometric scales composed of verbal questions, the Body Mass Index / Quetelet Index is an objective anthropometric measurement tool. Standardization of this instrument requires strict adherence to physical assessment protocols, precise instruments, and clear classification algorithms.

10.1. Measurement Protocol and Equipment Specifications

• Stadiometer (Height Measurement):
  • Equipment: Rigid vertical stadiometer with an adjustable horizontal headboard resting perpendicular to the vertical rule, readable to the nearest $0.1\text{ cm}$.
  • Protocol: Participant must be barefoot, wearing minimal hair accessories. Position the participant standing erect, heels together, with heels, buttocks, and upper back in contact with the vertical board. Position the head in the Frankfort Horizontal Plane (an imaginary line joining the orbitale—lower edge of the eye socket—to the tragion—superior notch of the ear canal). Ask the participant to take a deep inhalation and hold their breath while lowering the headboard firmly onto the crown of the head. Record height at peak inspiration.
• Medical Weighing Scale (Weight Measurement):
  • Equipment: Annually calibrated beam-balance scale or Class III digital medical electronic scale with a resolution of $0.1\text{ kg}$.
  • Protocol: Participant must void their bladder prior to measurement, remove shoes, heavy outerwear (coats, jackets), belts, and all items from pockets. The participant stands still, weight evenly distributed across both feet on the center of the platform. Record mass to the nearest $0.1\text{ kg}$.

10.2. Mathematical Calculation

Calculate the index using metric units:

$$\text{BMI } (\text{kg/m}^2) = \frac{\text{Weight in kilograms}}{(\text{Height in meters})^2}$$

For measurements recorded in Imperial units, use the conversion factor:

$$\text{BMI} = \frac{\text{Weight in pounds} \times 703}{(\text{Height in inches})^2}$$

10.3. Diagnostic Cut-offs and Classification Thresholds

The World Health Organization (WHO) establishes the international standard adult classification categories:

• Severe Thinness / Underweight Class III: $< 16.00\text{ kg/m}^2$ (Directly aligns with DSM-5 Severe to Extreme Anorexia Nervosa)
• Moderate Thinness / Underweight Class II: $16.00\text{–}16.99\text{ kg/m}^2$ (Aligns with DSM-5 Moderate Anorexia Nervosa)
• Mild Thinness / Underweight Class I: $17.00\text{–}18.49\text{ kg/m}^2$ (Aligns with DSM-5 Mild Anorexia Nervosa threshold)
• Normal / Healthy Weight Range: $18.50\text{–}24.99\text{ kg/m}^2$
• Pre-obesity / Overweight: $25.00\text{–}29.99\text{ kg/m}^2$
• Obesity Class I (Moderate): $30.00\text{–}34.99\text{ kg/m}^2$
• Obesity Class II (Severe): $35.00\text{–}39.99\text{ kg/m}^2$
• Obesity Class III (Very Severe / Morbid): $ge 40.00\text{ kg/m}^2$

10.4. Specialized Population Adjustments

• Asian Populations: Due to a higher percentage of visceral body fat and elevated cardiometabolic risk at lower BMI levels, the WHO Western Pacific Region recommends adjusted cut-offs: Overweight ($ge 23.0\text{ kg/m}^2$), Obesity Class I ($25.0\text{–}29.9\text{ kg/m}^2$), and Obesity Class II ($ge 30.0\text{ kg/m}^2$).
• Pediatric Populations (Ages 2–19): Raw BMI values cannot be interpreted using adult absolute thresholds due to dynamic growth trajectories. Instead, raw scores are converted into age- and sex-specific BMI-for-age percentiles or standard deviation z-scores based on CDC or WHO Growth Charts: Underweight ($< 5\text{th}$ percentile), Healthy weight ($5\text{th to } <85\text{th}$ percentile), Overweight ($85\text{th to } <95\text{th}$ percentile), and Obese ($ge 95\text{th}$ percentile or $\text{BMI} ge 30\text{ kg/m}^2$, whichever is lower).

11. Permissions, Fee, and Test Year

The Quetelet Index was first formally documented in 1842 in English translations of Quetelet’s treatise, following his initial presentations to the Royal Academy of Brussels in the 1830s. The modern clinical re-conceptualization as the Body Mass Index occurred in 1972 through the work of Ancel Keys and colleagues.

Because the index was developed in the 19th century and represents an objective mathematical formula based on fundamental physical units, it resides entirely within the public domain. There are no copyright protections, licensing constraints, or commercial fees associated with administering, scoring, implementing, or analyzing the Body Mass Index in clinical care, academic psychology, epidemiological research, or software applications.

12. References

The foundational, methodological, and clinical literature validating the Quetelet Index and the Body Mass Index includes the following works:

• American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.; DSM-5-TR). American Psychiatric Publishing. https://doi.org/10.1176/appi.books.9780890425787
• Eknoyan, G. (2008). Adolphe Quetelet (1796–1874)—the average man and indices of obesity. Nephrology Dialysis Transplantation, 23(1), 47–51. https://doi.org/10.1093/ndt/gfm517
• Flegal, K. M., Kit, B. K., Orpana, H., & Graubard, B. I. (2013). Association of all-cause mortality with overweight and obesity using standard body mass index categories: A systematic review and meta-analysis. JAMA, 309(1), 71–82. https://doi.org/10.1001/jama.2012.113905
• Global BMI Mortality Collaboration. (2016). Body-mass index and all-cause mortality: Individual-participant-data meta-analysis of 239 prospective studies in four continents. The Lancet, 388(10046), 776–786. https://doi.org/10.1016/S0140-6736(16)30175-1
• Keys, A., Fidanza, F., Karvonen, M. J., Kimura, N., & Taylor, H. L. (1972). Indices of relative weight and obesity. Journal of Chronic Diseases, 25(6–7), 329–343. https://doi.org/10.1016/0021-9681(72)90027-6
• Nuttall, F. Q. (2015). Body Mass Index: Obesity, BMI, and health: A critical review. Nutrition Today, 50(3), 117–128. https://doi.org/10.1097/NT.0000000000000092
• Quetelet, L. A. J. (1842). A treatise on man and the development of his faculties (R. Knox, Trans.). William and Robert Chambers.
• Stunkard, A. J., Sorensen, T., & Schulsinger, F. (1983). Use of the Danish Adoption Register for the study of obesity and thinness. Research Publications – Association for Research in Nervous and Mental Disease, 60, 115–120.
• World Health Organization. (2000). Obesity: Preventing and managing the global epidemic (WHO Technical Report Series, No. 894). World Health Organization. https://apps.who.int/iris/handle/10665/42330
• World Health Organization. (2004). Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. The Lancet, 363(9403), 157–163. https://doi.org/10.1016/S0140-6736(03)15268-3

13. Items of the Scale