With dietary intervention studies, it is important to ensure that subjects adhere to the test diet. Current methods to monitor adherence have substantial limitations. Therefore, a dose-response test curve was constructed to determine whether small differences in serum Li could be detected in response to ingestion of variable Li doses indicative of full or partial dietary compliance. During 3 separate weeks, subjects consumed a test meal that included a single food containing Li citrate daily for 4 d. Doses of 250, 213, or 175 micromol Li were used each week to approximate compliance levels of 100, 85, and 70%. On d 4, blood samples were taken before and 1, 2, 3, 5, 7, 9, and 24 h after ingesting the test meal. Compared with the 100% dose, serum Li was significantly lower at all times after the 70% dose and at most times after the 85% dose. Data were analyzed to determine a cutoff value so that if a subject's serum Li was below that value, they would be declared noncompliant. The probability that a subject was declared to be noncompliant when in fact they were compliant was set to 0.05 (specificity was set to 0.95) and the probability of noncompliance (sensitivity) was estimated. Test performance was best in the 3- to 9-h range, for which 90-95% of subjects "noncompliant" at the 70% dose were identified. Li can be used as a biomarker to determine dietary compliance. Measuring serum Li 3-9 h after the last dose provides the highest sensitivity and specificity for noncompliance at doses <70%.