Aim Controlled attenuation parameter (CAP) measured by vibration-controlled transient elastography (FibroScan) allows repeatable and reliable assessment of liver steatosis for screening of patients at risk of non-alcoholic steatohepatitis development among asymptomatic individuals at a community level. However, this has never been compared with another quantitative method, such as magnetic resonance imaging-based proton density fat fraction, among the Chinese health checkup population. Methods A multicenter prospective study was conducted with Chinese individuals undergoing regular health checkups. Steatosis grading by magnetic resonance imaging-based proton density fat fraction was used as the reference to evaluate the diagnostic performance of CAP. Results A total of 173 individuals were included with mean age of 45 +/- 11 years and body mass index of 25.8 +/- 4.0 kg/m(2). A linear correlation was found between CAP and log(10)-transformed magnetic resonance imaging-based proton density fat fraction results (Pearson's coefficient 0.772, P < 0.001). The areas under the receiver operating characteristic curve for distinguishing >= S1 and >= S2 steatosis were 0.88 (95% confidence interval 0.83-0.93) and 0.89 (95% confidence interval 0.83-0.95), respectively. When optimized for >= 90% sensitivity, the CAP cut-off for staging >= S1 steatosis was 244 dB/m. CAP could classify patients with >= S1 steatosis with similar performance as an ultrasound examination. Conclusions As a non-invasive and quantitative method, CAP is highly adapted for population screening at a community level. With the integration of liver stiffness and CAP results in risk stratification scores for non-alcoholic steatohepatitis, vibration-controlled transient elastography can be useful in regular health checkups.