OBJECTIVES: To test the hypothesize that increased heart rate (HR) and decreased heart rate variability (HRV) are not only due to concurrent stressful events and worries but also to stressors and worries occurring in the preceding hours or stressors anticipated to occur in the next hour. Worry was expected to mediate at least part of the prolonged effects of stressors. METHODS: Ambulatory HR and HRV of 73 teachers were recorded for 4 days, during which the participants reported occurrence and duration of worry episodes and stressful events on an hourly basis, using computerized diaries. Multilevel regression models were used, accounting for effects of several biobehavioral variables. RESULTS: Stressful events were not associated with changes in HR or HRV. However, worry episodes had effects on concurrent HR and HRV (2.55 beats/minute; -5.76 milliseconds) and HR and HRV in the succeeding hour (3.05 beats/minute; -5.80 milliseconds) and 2 hours later (1.52 beats/minute; -3.14 milliseconds). These findings were independent of emotions, physical activity, posture, and other biobehavioral factors. CONCLUSION: Worry has effects on cardiac activity, and these effects were still visible after 2 hours. The latter finding suggests that a considerable part of prolonged activation may be induced by unconscious stress-related cognition.