It has been reported that the frequency of cola intake (COLA) is positively associated with serum triglycerides and negatively associated with high-density-lioprotein (HDL) cholesterol, both components of the metabolic syndrome (MetS). The question now is whether noncola soft drink intake (NCOLA) is associated with MetS. Among the 18 770 participants in the Oslo Health Study, 5373 men and 6181 women had data on COLA and NCOLA and risk factors for MetS (except fasting glucose). Main MetS requirements are central obesity and 2 of the following: increased triglycerides, low HDL cholesterol, increased systolic or diastolic blood pressure, and elevated fasting blood glucose. The MetSRisk index was calculated to estimate many MetS components. Using regression analyses, the association between COLA (NCOLA) and MetS (MetSRisk) was studied. In young (aged 30 years), middle-aged (aged 40 and 45 years), and senior (aged 59 and 60 years) men and women, there was, in general, a positive correlation between COLA and MetSRisk, and between COLA and single MetS risk factors, except HDL cholesterol, which was negatively correlated. A less consistent picture was found for NCOLA. By regression analyses, after adjustment for sex, age, time since last meal, and use of sugar-sweetened soft drinks, a positive association between COLA (NCOLA) and MetSRisk (MetS) was still found. However, when also controlling for cheese, fatty fish, coffee, alcohol, smoking, physical activity, education, and birthplace, only the association with COLA remained significant, irrespective of the presence or absence of sugar. In conclusion, the self-reported intake frequency of soft drinks can be positively associated with MetS.