The discriminatory ability of the body roundness index and body mass index for metabolic diseases in Korean adults: a comparative study

Classified as probably not a substantial claim.

Several studies have shown that BRI demonstrates superior screening performance compared with traditional indices such as BMI, waist circumference, and waist-to-height ratio, particularly for metabolic diseases including metabolic syndrome.

“Several studies have shown that BRI demonstrates superior screening performance compared with traditional indices such as BMI, waist circumference, and waist-to-height ratio, particularly for metabolic diseases, including metabolic syndrome [12,13].”

• cites [12] sourcePartly supported

  The Conclusions say BRI was a superior predictor to BMI, waist circumference, and waist-to-height ratio for a cluster of cardiometabolic abnormalities including metabolic syndrome, especially in Chinese women. But the Discussion (Body roundness index) says BRI and waist-to-height ratio had coincident predictive ability, and the Introduction reports a prior study in which BRI was not superior to BMI or waist circumference. This source therefore supports the narrower finding from its own study, not the unqualified claim that several studies found BRI superior across these comparisons. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [13] sourceNot checked — no full text of the source

May combine two claims.

Mousavi et al. reported that during reproductive age, both BMI and BRI were significantly higher in female with diabetes than in those without, whereas after menopause, only BRI remained significantly different.

“Mousavi et al. [27], in a study of diabetes risk across the female life cycle, compared premenopausal and postmenopausal female and reported that during reproductive age, both BMI and BRI were significantly higher in female with diabetes than in those without, whereas after menopause, only BRI remained significantly different.”

• cites [27] sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Previous studies reported a U-shaped association between BRI and mortality similar to that observed for BMI, with both low and high BRI levels linked to increased all-cause mortality.

“One study evaluating the association between BRI and mortality identified a U-shaped curve similar to that observed for BMI [14].”

• Also stated in: § DISCUSSION · plain

  “Previous studies have also examined the relationship between BRI and mortality, reporting a U-shaped association similar to that observed for BMI, where both low and high BRI levels were linked to increased all-cause mortality [14,25].”

• cites [14] sourcePartly supported

  The Discussion reports a U-shaped association between BRI and all-cause mortality, with increased risk at both low and high BRI levels. The Results > Sensitivity Analyses section compares BRI with BMI but does not state that BMI had a similar U-shaped curve. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [25] sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Yang et al. reported an increased risk of cardiovascular disease according to BRI trajectory, suggesting that BRI may serve as a novel indicator for cardiovascular disease risk.

“In addition, Yang et al. [15] reported an increased risk of cardiovascular disease according to BRI trajectory, suggesting that BRI may serve as a novel indicator for cardiovascular disease risk.”

• cites [15] sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

KNHANES is a nationwide, representative, cross-sectional survey conducted annually by the Korea Disease Control and Prevention Agency since 1998.

“KNHANES is a nationwide, representative, cross-sectional survey conducted annually by the Korea Disease Control and Prevention Agency (KDCA) since 1998 in accordance with the National Health Promotion Act.”

• cites [16] sourcePartly supported

  The Introduction states that KNHANES began in 1998, represents Koreans, and is conducted annually; Strengths and Weaknesses identifies its cross-sectional design. Our copy does not state that KDCA has conducted the survey since 1998 or that it is conducted under the National Health Promotion Act. It identifies a KDCA KNHANES website, but that does not establish those additional details. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Elffers et al. suggested that postmenopausal female may experience more adverse cardiometabolic consequences of visceral obesity compared with male of similar age.

“Elffers et al. [26] suggested that postmenopausal female may experience more adverse cardiometabolic consequences of visceral obesity compared with male of similar age.”

• cites [26] sourcePartly supported

  The Discussion (page unavailable) reports a stronger association between visceral fat and cardiometabolic risk factors in obese women than in obese men, and notes that women's fat distribution shifts toward visceral fat after menopause. It does not report a comparison specifically between postmenopausal women and men of similar age, so that qualifier is not established. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

BRI was developed by Thomas et al. in 2013 based on fat mass measured by dual-energy X-ray absorptiometry and visceral fat measured by magnetic resonance imaging, along with waist circumference and height.

“BRI was developed by Thomas et al. [11] in 2013 based on fat mass measured by dual-energy X-ray absorptiometry and visceral fat measured by magnetic resonance imaging, along with waist circumference and height.”

• cites [11] sourcePartly supported

  The Elliptical Models and Body Roundness Index sections describe BRI as a transformation of an ellipse’s eccentricity, calculated using height and waist circumference. The Subjects and Design and Methods sections report DXA-measured body fat and MRI-measured visceral fat, but these measurements were used to develop and assess prediction models, not to calculate BRI itself. The claim conflates those roles. (LLM, conf=0.87)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Suliga et al. found that for metabolic syndrome, BMI performed better than BRI in male.

“Similarly, Suliga et al. [24] found that for metabolic syndrome, BMI performed better than BRI in male.”

• cites [24] sourcePartly supported

  The Discussion (page not available) ranks BMI among the stronger predictors for men under the modified metabolic-syndrome definition, while BRI is not among those listed. The claim omits that definition and presents the comparison as applying to metabolic syndrome generally; the passages do not establish BMI’s advantage over BRI under the standard definition. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

An analysis based on the Korean National Health Insurance Service reported that waist circumference was associated with increased risks of type 2 diabetes, hypertension, and myocardial infarction, and that these risks further increased with higher BMI values.

“In the Korean context, an analysis based on the Korean National Health Insurance Service (NHIS) reported that waist circumference, as a measure of abdominal obesity, was associated with increased risks of type 2 diabetes, hypertension, and myocardial infarction, and that these risks further increased with higher BMI values [22].”

• cites [22] sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Maessen et al. reported that although BRI demonstrated discriminatory ability for cardiovascular diseases, it was not superior to BMI as an anthropometric measure.

“Maessen et al. [23] reported that although BRI demonstrated discriminatory ability for cardiovascular diseases, it was not superior to BMI as an anthropometric measure.”

• cites [23] sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Wu et al. found that BRI was strongly associated with cardiovascular disease in younger individuals aged 55 years and below.

“Wu et al. [21] also found that BRI was strongly associated with cardiovascular disease in younger individuals aged 55 years and below.”

• cites [21] sourcePartly supported

  The Results report that the association between BRI trajectories and CVD was more evident in adults aged <55 years. The claim broadly reflects that finding, but it says BRI rather than BRI trajectories and includes people aged 55, whereas the paper's younger subgroup excludes them. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Tian et al. found that BRI demonstrated superior discriminatory ability compared with BMI for all metabolic conditions in female and for hypertension and diabetes in male.

“Tian et al. [12], using data from the 2008-2009 China Health and Nutrition Survey, analyzed AUROC values for several metabolic diseases and found that BRI demonstrated superior discriminatory ability compared with BMI for all metabolic conditions (hypertension, diabetes, metabolic syndrome, and dyslipidemia) in female, and for hypertension and diabetes in male.”

• cites [12] sourceSupported

May combine two claims.

For all metabolic diseases, ORs of both BMI and BRI were highest in younger age groups and decreased with increasing age, consistent in both male and female.

“For all metabolic diseases, the ORs of both BMI and BRI were highest in the younger age groups and decreased with increasing age, and this pattern was consistent in both male and female (Table 3).”

May combine two claims.

For all conditions, ORs for BRI exceeded those for BMI, with the largest differences for diabetes and metabolic syndrome, where ORs were approximately 1.9-fold and 2.4-fold higher, respectively.

“For all conditions, ORs for BRI exceeded those for BMI, with the largest differences observed for diabetes and metabolic syndrome, in which ORs were approximately 1.9-fold and 2.4-fold higher, respectively.”

• Also stated in: § DISCUSSION · plain

  “Additionally, logistic regression analysis showed that the highest quintiles of BRI were associated with substantially greater risks of metabolic diseases compared with BMI, with differences of approximately 2-fold in magnitude for diabetes and metabolic syndrome.”

May combine two claims.

Across diabetes, hypertension, hypercholesterolemia, and metabolic syndrome, AUROC values ranged from 0.739 to 0.844 for BMI and from 0.745 to 0.851 for BRI.

“AUROC values ranged from 0.739 to 0.844 for BMI and 0.745 to 0.851 for BRI, with both indices demonstrating their highest performance for metabolic syndrome.”

• Also stated in: § Area under the receiver operating characteristic curves of body mass index and body roundness index for metabolic disease · plain

  “The AUROC values of BMI and BRI for diabetes, hypertension, hypercholesterolemia, and metabolic syndrome ranged from 0.739 to 0.844 and from 0.745 to 0.851, respectively.”

May combine two claims.

For all four metabolic conditions, both BMI and BRI were significantly higher in the disease group than in the non-disease group.

“For all 4 conditions, both BMI and BRI were significantly higher in the disease group compared with the non-disease group (Table 1).”

May combine two claims.

The most pronounced AUROC differences between BRI and BMI were for diabetes in female aged 45 years or older and for metabolic syndrome in male aged 45 years or older.

“The most pronounced differences between the 2 indices were noted for diabetes in female aged 45 years or older and for metabolic syndrome in male aged 45 years or older (Table 2).”

• Also stated in: § RESULTS · plain

  “Subgroup analyses identified the most pronounced differences for diabetes in female aged 45 years or older and for metabolic syndrome in male aged 45 years or older.”

May combine two claims.

Both BMI and BRI showed dose-response relationships with metabolic diseases, with ORs increasing across higher quintiles, and this pattern was consistent regardless of sex or age.

“Both indices demonstrated clear dose-response relationships, with ORs for all metabolic diseases increasing steadily across higher quintiles, and this pattern remained consistent regardless of sex or age (Figure 3).”

May combine two claims.

BRI AUROC was significantly higher than BMI AUROC for diabetes and metabolic syndrome, but not for hypertension or hypercholesterolemia.

“The AUROC values of BRI were significantly higher than those of BMI for diabetes and metabolic syndrome, whereas no significant differences were found for hypertension and hypercholesterolemia.”

May combine two claims.

In sex-specific OR analysis, BMI had higher ORs in male than female for all conditions except diabetes, whereas BRI had higher ORs in female than male for all conditions except hypercholesterolemia.

“In the sex-specific analysis, BMI showed higher ORs in male than in female for all conditions except diabetes, whereas BRI showed higher ORs in female than in male for all conditions except hypercholesterolemia.”

May combine two claims.

The BRI-BMI AUROC difference was greatest for metabolic syndrome in male (0.015) and for diabetes in female (0.011).

“The difference between BRI and BMI was greatest for metabolic syndrome in male (0.015), whereas in female, the largest difference was observed for diabetes (0.011).”

May combine two claims.

In the total study population, BRI had higher AUROC values than BMI for diabetes, hypercholesterolemia, and metabolic syndrome, but not for hypertension, with the largest difference observed for diabetes.

“BRI outperformed BMI for 3 metabolic diseases except hypertension, with the largest AUROC difference observed for diabetes (0.01).”

• Also stated in: § Area under the receiver operating characteristic curves of body mass index and body roundness index for metabolic disease · plain

  “In the total study population, the AUROC values of BRI were higher than those of BMI for 3 metabolic diseases, except hypertension (Figure 2).”

• Also stated in: § DISCUSSION · boosted

  “BRI demonstrated better discriminatory ability than BMI for all metabolic diseases except hypertension, with the greatest difference observed for diabetes.”

May combine two claims.

For both BMI and BRI, the highest AUROC was observed for metabolic syndrome and the lowest for hypercholesterolemia.

“For both indices, the highest AUROC was observed for metabolic syndrome, and the lowest for hypercholesterolemia.”

May combine two claims.

BRI more accurately reflects central obesity and age-related changes in body shape because it incorporates both waist circumference and height into a geometric model.

“In contrast, BRI, which incorporates both waist circumference and height into a geometric model, more accurately reflects central obesity and age-related changes in body shape.”

May combine two claims.

The decline in ORs with advancing age suggests that the predictive power of anthropometric indices diminishes in older populations, likely due to age-related changes in body composition and potential survivorship bias.

“By contrast, the decline in ORs with advancing age suggests that the predictive power of anthropometric indices diminishes in older populations, likely due to age-related changes in body composition and potential survivorship bias.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The larger differences between BRI and BMI in female compared with male may be partly attributed to sex-specific differences in fat distribution and hormonal changes after menopause.

“The larger differences between BRI and BMI seen in female compared with male may be partly attributed to sex-specific differences in fat distribution and hormonal changes after menopause, which intensify the metabolic consequences of visceral obesity.”

Classified as probably not a substantial claim.

All ROC models were based on multivariable logistic regression analyses adjusted for age, sex, education level, household income, physical activity, smoking status, and alcohol consumption.

“All ROC models were based on multivariable logistic regression analyses adjusted for age, sex, education level, household income, physical activity, smoking status, and alcohol consumption.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Our findings are consistent with previous international research.

“Our findings are consistent with previous international research.”

May combine two claims.

Discrepancies among studies may be explained by differences in study populations, cutoff values, and disease outcomes, suggesting that the relative performance of BRI and BMI may depend on population-specific and methodological factors.

“These discrepancies may be explained by differences in study populations, variations in cutoff values, and heterogeneity in disease outcomes assessed, suggesting that the relative performance of BRI and BMI may depend heavily on population-specific and methodological factors.”

May combine two claims.

BMI does not differentiate fat from muscle mass or reflect fat distribution, which limits its usefulness in disease screening.

“However, it does not differentiate fat from muscle mass or reflect fat distribution, which limits its usefulness in disease screening [6].”

• cites [6] sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

BRI could serve as a complementary indicator to BMI, offering additional value in clinical practice and national health monitoring systems.

“These findings suggest that BRI could serve as a complementary indicator to BMI, offering additional value in clinical practice and national health monitoring systems.”

May combine two claims.

The association between BRI and metabolic syndrome may be inflated, and the observed OR magnitude may not reflect causal strength but rather the shared definitional component.

“As a result, the association between BRI and metabolic syndrome may be inflated, and the magnitude of the observed ORs may not reflect causal strength but rather the shared definitional component.”

Probably not a substantial claim: may not pass S3 (load-bearing).

All analyses incorporated sampling weights, clusters, and strata to account for the complex survey design of KNHANES, except for ROC analyses.

“All analyses incorporated sampling weights, clusters, and strata to account for the complex survey design of KNHANES, except for ROC analyses, as the purpose of ROC analysis was to compare the discriminatory ability of BMI and BRI rather than to estimate population-level parameters.”

Classified as probably not a substantial claim.

Logistic regression analysis for ORs was adjusted for age (continuous), sex, education level, household income, physical activity, smoking status, and alcohol consumption.

“Logistic regression analysis was conducted to calculate odds ratios (ORs) for each higher group relative to the lowest group, adjusting for age (continuous), sex, education level, household income, physical activity, smoking status, and alcohol consumption.”

Classified as probably not a substantial claim.

The comparison between BMI and BRI was considered fair because both indices were analyzed under the same conditions.

“Nevertheless, the comparison between BMI and BRI was considered fair, as both indices were analyzed under the same conditions.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Because this study employed a cross-sectional design, causal relationships between variables cannot be established.

“First, because this study employed a cross-sectional design, causal relationships between variables cannot be established.”

May combine two claims.

Several confounding variables could not be controlled for: family history variables were excluded due to substantial missing data, and dietary habits or nutritional status variables were removed because of the possibility of reverse causation.

“Second, several confounding variables could not be controlled for: variables related to family history were excluded due to substantial missing data, and those related to dietary habits or nutritional status were removed because of the possibility of reverse causation.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The use of nationally representative data minimized sampling error and strengthened statistical precision.

“The use of nationally representative data minimized sampling error and strengthened statistical precision.”