BACKGROUND: The rising prevalence of obesity has made hepatic steatosis an increasingly common issue. Ultrasound is generally used in clinical practice to assess steatosis, but its accuracy has been inconsistent across studies. We aimed to determine the validity of ultrasound to diagnose hepatic steatosis when compared to the criterion method proton magnetic resonance spectroscopy (MRS) in older individuals. METHODS: A total of 72 healthy white European individuals (n = 42 men; n = 30 women aged 67-76 years) participating in the Hertfordshire Birth Cohort Physical Activity trial had hepatic steatosis assessed by ultrasound and MRS. The ultrasound scans were graded as normal, mild, moderate and severe steatosis, while hepatic fat content above 5.5% by MRS was used as a cut-off for steatosis. RESULTS: 18 participants (25%) had a level of hepatic fat measured by MRS consistent with diagnosis of steatosis. The sensitivity and specificity of ultrasound in diagnosing hepatic steatosis (mild/moderate/severe vs normal) were 96% (95% CI: 87-99.6%) and 94% (95% CI: 73-100%) respectively, although overlap in MRS hepatic fat content was observed between the ultrasound categories. CONCLUSIONS: Ultrasound is a valid method for detecting the presence or absence of hepatic steatosis in older adults and can be used as an alternative tool in both clinical investigations and epidemiological studies, when other imaging techniques are not feasible.