Dataset Information

Generalized lambda distribution for flexibly testing differences beyond the mean in the distribution of a dependent variable such as body mass index.

ABSTRACT: BACKGROUND/OBJECTIVES:Conventional statistical methods often test for group differences in a single parameter of a distribution, usually the conditional mean (for example, differences in mean body mass index (BMI; kg?m-2) by education category) under specific distributional assumptions. However, parameters other than the mean may of be interest, and the distributional assumptions of conventional statistical methods may be violated in some situations. SUBJECTS/METHODS:We describe an application of the generalized lambda distribution (GLD), a flexible distribution that can be used to model continuous outcomes, and simultaneously describe a likelihood ratio test for differences in multiple distribution parameters, including measures of central tendency, dispersion, asymmetry and steepness. We demonstrate the value of our approach by testing for differences in multiple parameters of the BMI distribution by education category using the Health and Retirement Study data set. RESULTS:Our proposed method indicated that at least one parameter of the BMI distribution differed by education category in both the complete data set (N=13?571) (P<0.001) and a randomly resampled data set (N=300 from each category) to assess the method under circumstances of lesser power (P=0.044). Similar method using normal distribution alternative to GLD indicated the significant difference among the complete data set (P<0.001) but not in the smaller randomly resampled data set (P=0.968). Moreover, the proposed method allowed us to specify which parameters of the BMI distribution significantly differed by education category for both the complete and the random subsample, respectively. CONCLUSIONS:Our method provides a flexible statistical approach to compare the entire distribution of variables of interest, which can be a supplement to conventional approaches that frequently require unmet assumptions and focus only on a single parameter of distribution.

SUBMITTER: Ejima K

PROVIDER: S-EPMC5927854 | biostudies-literature | 2018 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Generalized lambda distribution for flexibly testing differences beyond the mean in the distribution of a dependent variable such as body mass index.

Ejima K K Pavela G G Li P P Allison D B DB

International journal of obesity (2005) 20171030 4

<h4>Background/objectives</h4>Conventional statistical methods often test for group differences in a single parameter of a distribution, usually the conditional mean (for example, differences in mean body mass index (BMI; kg m<sup>-2</sup>) by education category) under specific distributional assumptions. However, parameters other than the mean may of be interest, and the distributional assumptions of conventional statistical methods may be violated in some situations.<h4>Subjects/methods</h4>We ...[more]

PMID: 29081503

Similar Datasets

Project description:The clinical recognition of cardiometabolic disorders might be enhanced by anthropometry based on the sagittal abdominal diameter (SAD; also called "abdominal height") or waist circumference rather than on weight. Direct comparisons of body mass index (BMI, weight/height(2)) with SAD/height ratio (SADHtR) or waist circumference/height ratio (WHtR) have not previously been tested in nationally representative populations.Nonpregnant adults without diagnosed diabetes (ages 20-64 years; n = 3071) provided conventional anthropometry and supine SAD (by sliding-beam caliper) in the 2011-2012 US National Health and Nutrition Examination Survey. Population-weighted, logistic models estimated how strongly each anthropometric indicator was associated with 5 cardiometabolic disorders: Dysglycemia (glycated hemoglobin ?5.7%), HyperNonHDLc (non-high-density-lipoprotein [HDL] cholesterol ?4.14 mmol/L, or taking anticholesteremic medications), Hypertension (systolic blood pressure ?140 mm Hg or diastolic blood pressure ?90 mm Hg, or taking antihypertensive medications), HyperALT (alanine transaminase ?p75 [75th percentile, sex-specific]), and HyperGGT (gamma-glutamyltransferase ?p75 [sex-specific]).After scaling each indicator, adjusted odds ratios (aORs) tended to be highest for SADHtR and lowest for BMI when identifying each disorder except dysglycemia. When SADHtR entered models simultaneously with BMI, the aORs for BMI no longer directly identified any condition, whereas SADHtR identified persons with HyperNonHDLc by aOR 2.78 (95% confidence interval [CI], 1.71-4.51), Hypertension by aOR 2.51 (95% CI, 1.22-5.15), HyperALT by aOR 2.89 (95% CI, 1.56-5.37), and HyperGGT by aOR 5.43 (95% CI, 3.01-9.79). WHtR competed successfully against BMI with regard to Dysglycemia, HyperNonHDLc, and HyperGGT. c-Statistics of SADHtR and WHtR were higher than those of BMI (P <.001) for identifying HyperNonHDLc and HyperGGT.Among nonelderly adults, SADHtR or WHtR recognized cardiometabolic disorders better than did the BMI.

Project description:Highlights • Women with patellofemoral pain have higher body mass index and body fat and lower skeletal muscle mass compared to pain-free women.• Men with patellofemoral pain and both sexes combined have no difference in body mass index, body fat, or skeletal muscle mass compared to their respective pain-free control groups.• Body mass index and levels of body fat and skeletal muscle mass should be considered when assessing women with patellofemoral pain.• Future studies should not assess body composition measures in a mixed-sex population without discriminating between men and women. Background We compared body mass index (BMI), body fat, and skeletal muscle mass between (1) a mixed-sex nonathletic cohort of people with patellofemoral pain (PFP) and pain-free people, and (2) a nonathletic cohort of people with PFP and pain-free people subgrouped by sex (i.e., men and women with PFP vs. pain-free men and women). Methods This cross-sectional study included 114 people with PFP (71 women, 43 men) and 54 pain-free controls (32 women, 22 men). All participants attended a single testing session to assess body composition measures, which included BMI, percentage of body fat (%BFBioimpedance), and skeletal muscle mass (both assessed by bioelectrical impedance analysis), and percentage of body fat (%BFSkinfold) (assessed by skinfold caliper analysis). A one-way univariate analysis of covariance (age and physical activity levels as covariates) was used to compare body composition measures between groups (i.e., PFP vs. pain-free group; women with PFP vs. pain-free women; men with PFP vs. pain-free men). Results Women with PFP presented significantly higher BMI, %BFBioimpedance, and %BFSkinfold, and lower skeletal muscle mass compared to pain-free women (p ≤ 0.04; effect size : ‒0.47 to 0.85). Men with PFP and men and women combined had no differences in BMI, %BFBioimpedance, %BFSkinfold, and skeletal muscle mass compared to their respective pain-free groups (p > 0.05). Conclusion Our findings indicate that BMI and body composition measures should be considered as part of the evaluation and management of people with PFP, especially in women, who have demonstrated higher BMI and body fat and lower skeletal muscle mass compared to pain-free controls. Future studies should not assess body composition measures in a mixed-sex population without distinguishing men participants from women participants. Graphical abstract Image, graphical abstract

Dataset Information

Generalized lambda distribution for flexibly testing differences beyond the mean in the distribution of a dependent variable such as body mass index.

Publications

Generalized lambda distribution for flexibly testing differences beyond the mean in the distribution of a dependent variable such as body mass index.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets