Nutritional epidemiology aims to link dietary exposures to chronic disease, but instruments for evaluating dietary intake are inaccurate (termed here dietary misreporting). Identifying unreliable data and the sources of errors are key goals. This can be achieved by comparing estimated intakes to total energy expenditure (TEE). Here we use the IAEA doubly-labeled water database to derive a predictive equation for TEE using 6497 measures of TEE in individuals aged 4 to 96 years old. The resultant regression equation predicts expected TEE from easily acquired variables, like body weight, age and sex, with 95% predictive limits that can be used to screen for misreporting by participants in dietary studies. We applied the equation to two large data sets (NDNS and NHANES) and found the level of misreporting was >50%. Macronutrient composition from dietary reports in these studies was systematically biased as the level of misreporting increased, leading to potentially spurious associations between diet components and body mass index (BMI). This issue can be avoided using the screening method derived here.