Nutrition/Obesity
Using signal detection analysis to identify subgroups of midlife women in Mexico City with differing prevalence of metabolic syndrome Yanxin Zhu* Yanxin Zhu Zhu Zhu Zhu Zhu Zhu Zhu Zhu University of Michigan School of Public Health
Objective: To identify heterogeneous subgroups of midlife women in Mexico City with varying prevalence of metabolic syndrome (MetS) using signal detection analysis (SDA).
Methods: This cross-sectional study included 443 women from the 2019-2023 visit of the Early Life Exposure in Mexico to ENvironmental Toxicants (ELEMENT) cohort. MetS was defined according to the National Cholesterol Education Program Adult Treatment Panel III criteria. Factors potentially associated with MetS were selected based on the socioecological model and included demographic characteristics, socioeconomic position, health behaviors, body mass index (BMI), and menopause-related measures. Data processing was conducted in R 4.4.2, and SDA was performed using ROC5 software.
Results: MetS prevalence was 49.0% in this population of midlife women (mean age: 48.4±6.3 years). SDA identified 8 mutually exclusive subgroups, with MetS prevalence ranging from 10.9% to 78.8%. BMI was the primary splitting variable, with additional differentiation by the number of children, age, sedentary time, moderate-to-vigorous physical activity, years of education, ultra-processed food (UPF), and total energy intake. The highest prevalence group (78.8%) was characterized by BMI≥30 kg/m2, <11 years of education, less than 17.6% grams/day from UPF intake, and sedentary time <11.5 hr/day. The lowest prevalence group (10.9%) was defined as BMI<25 kg/m2 and having fewer than 3 children.
Conclusion: Distinct combinations of BMI, age, socioeconomic status, diet, physical activity, and sedentary behavior distinguished subgroups with markedly different MetS prevalence among midlife women from Mexico City. These findings highlight the importance of considering intersecting risk profiles rather than single factors when identifying populations at elevated metabolic risk. Longitudinal research is needed to determine whether these subgroup patterns predict the progression of metabolic risk.
