Dietary patterns derived from reduced rank regression and their associations with metabolic syndrome in middle-aged and older adults: A cross-sectional study in Selangor, Malaysia

Metabolic syndrome (MetS) is a rising public health concern in Malaysia, with a 35.9% adult prevalence reported by the 2023 National Health and Morbidity Survey. This cross-sectional study, based on secondary data analysis, aimed to investigate the association between a dietary pattern (DP) derived using reduced rank regression (RRR) and MetS among Malaysian adults aged 45 years and older. A total of 230 participants (89 men, 141 women) were recruited via stratified random sampling from Puchong, Serdang, and Kajang. The ethnic composition of the sample was 59% Chinese, 22% Malay, and 19% Indian. Dietary data were collected using a validated food frequency questionnaire, and MetS was defined according to the Joint Interim Statement (JIS) criteria. Thirty-one percent of participants were identified as having MetS. RRR identified a DP characterised by high intakes of total fat, fibre, and percentage sugar but low in dietary energy density, explaining 58% of the variation in metabolic risk response variables. Multivariable regression models, adjusted for age, gender, ethnicity, and smoking status were used to examine associations between DP scores, MetS, and its individual components. Among women, each 1-SD increase in adherence to the identified DP was associated with higher body mass index (β = 0.061; 95% CI: 0.000–0.123), greater waist circumference (β = 0.049; 95% CI: 0.012–0.086), a three-fold increase in the odds of elevated fasting blood glucose (OR = 3.23; 95% CI: 1.07–9.79), and reduced odds of low HDL-C (OR = 0.27; 95% CI: 0.08–0.97). No significant associations were observed among men. The identified DP defined by high consumption of fat, fibre, and free sugars, coupled with low energy density, was paradoxically associated with both risk-enhancing and risk-reducing metabolic parameters. The pattern was positively associated with hyperglycaemia and central obesity, while demonstrating an inverse association with low HDL-C concentrations.