1 Department of Mathematics and Statistics, Concordia University, Montreal, Quebec, Canada; PERFORM Centre, Concordia University, Montreal, Quebec, Canada. Electronic address: lisa.kakinami@concordia.ca.
2 Department of Chemistry, Concordia University, Montreal, Quebec, Canada.
3 Faculty of Land and Food Systems, Food, Nutrition and Health, University of British Columbia, Vancouver, British Columbia, Canada.
4 PERFORM Centre, Concordia University, Montreal, Quebec, Canada; Department of Health, Kinesiology, and Applied Physiology, Concordia University, Montreal, Quebec, Canada; Metabolism, Obesity, Nutrition Lab, PERFORM Centre, Concordia University, Montreal, Quebec, Canada.
5 Department of Health, Kinesiology, and Applied Physiology, Concordia University, Montreal, Quebec, Canada; Metabolism, Obesity, Nutrition Lab, PERFORM Centre, Concordia University, Montreal, Quebec, Canada.

Body mass index is poor at distinguishing between adiposity and muscle. Based on dual energy X-ray absorptiometry data, a diagnostic framework to analyze body composition by categorizing fat- and muscle-mass body composition into four phenotypes has been proposed. The objective of this study was to assess the association between body-composition phenotypes with adiposity measures, health behaviours and cardiometabolic risks in a representative U.S. adult population. Data were from NHANES (1999-2006: n = 9867; 2011-2018: n = 10,454). Four phenotypes based on being above/below the 50th percentile of age- and sex- adjusted reference curves of fat-mass and muscle-mass were identified. Multiple linear and logistic regressions were used to assess phenotypes (high [H] or low [L] adiposity [A] or muscle mass [M]) against adiposity measures, health behaviours, cardiometabolic risk, and dietary intake. Low-adiposity/high-muscle (LA-HM) was the referent. Analyses incorporated the complex sampling design and survey weights, and were adjusted for age, sex, race, and education. Compared to the LA-HM reference group, the HA-LM phenotype was less physically active, had higher total and lower high-density lipoprotein cholesterol, and had lower intake of all examined nutrients (all p < 0.01). For the HA-HM phenotype, unfavourable values were detected for all adiposity and cardiometabolic measures compared to the LA-HM phenotype (all p < 0.01). The two high adiposity phenotypes were associated with poorer health behaviours and cardiovascular risk factors, regardless of muscle-mass, but associations differed across the phenotypes. Results further underscores the importance of accounting for both adiposity and muscle mass in measurement and analysis. Further longitudinal investigation is needed.