Introduction: Although the majority of sodium intake is estimated to come from commercially processed/restaurant foods, about 11% is estimated to come from discretionary salt added at the table or during home cooking/preparation. Marked changes in U.S. food habits/choices, such as eating out, as well as the demographic composition of the population could change the frequency of discretionary salt use and invalidate past estimates. Objectives: To evaluate U.S. temporal trends in, and demographic/health determinants of, self-reported frequency of discretionary salt use (excluding 4% of population who use salt substitutes or lite salt). Methods: We analyzed salt intake questions for 31,842 persons aged ≥2y from the 2003-10 National Health and Nutrition Examination Survey (NHANES). We used multiple logistic regression models to assess temporal trends in reported frequency of discretionary salt use from 2003-10, adjusting for age, sex and race/ethnicity. We used chi-square tests to assess current (2007-10) differences in discretionary salt use by demographic/health characteristics. Analyses were adjusted for complex sampling design. Results: Using salt “very often” at the table declined from 2003-04 to 2009-10 (18% to 14%, p<0.01 for trend). The percent decline was greatest among male adults aged ≥19y (21% to 16%, p<0.01). Using salt “very often” during home cooking/preparation decreased (41% to 37%, p=0.03), while using salt “occasionally” increased (34% to 37%, p=0.04). Temporal trends in “never” and “rarely” using salt at the table or during home cooking/preparation were not statistically significant. In the 2 recent cycles of NHANES (2007-10), 33% of persons aged ≥2y reported “never” added salt at the table, 31% “rarely,” 22% “occasionally” and 15% “very often.” Corresponding percentages for frequency of salt added during home cooking/preparation were 7%, 19%, 36% and 37%. Overall, being non-Hispanic black, lower income and self-reported hypertension or diabetes were associated with being more likely to report never adding salt at the table or during cooking/preparation (all p<0.01). Among adults, as age increased, the percentage reporting “never” using salt increased both at the table (23% for ages 19-30y to 43% for ≥71y) and during home cooking/preparation (5% to 14%). The opposite was observed for children/adolescents, as age increased from 2-18y, the percentage reporting “never” using salt decreased at the table (79% for ages 2-3y to 18% for 14-18y) and during home cooking/preparation (6% to 2.5%). Conclusions: Frequency of high levels of discretionary salt use decreased from 2003-10. In general, persons with higher risk of elevated blood pressure were more likely to report using discretionary salt less frequently. The association of discretionary salt use with food habits/choices and the amount of sodium intake from discretionary salt merits further investigation.