OBJECTIVE: To test whether changes in the patient activation measure (PAM) are related to changes in health status and healthy behaviors. METHODS: Data for this secondary analysis were taken from a group-randomized, controlled trial comparing a traditional health promotion program for employees with an activated consumer program and a control program. The study population included 320 employees (with and without chronic disease) from two U.S. companies: a large, integrated health care system and a national airline. Survey and biometric data were collected in Spring 2005 (baseline) and Spring 2007 (follow-up). RESULTS: Change in PAM was associated with changes in health behaviors at every level (1-4), especially at level 4. Changes related to overall risk score and many of its components: aerobic exercise, safety, cancer risk, stress and mental health. Other changes included frequency of eating breakfast and the likelihood of knowing about health plans and how they compare. CONCLUSION: Level 4 of patient activation is not an end-point. People are capable of continuing to make significant change within this level. PRACTICE IMPLICATIONS: Interventions should be designed to encourage movement from lower to higher levels of activation. Even people at the most activated level improve health behaviors.