The aim of this study was to assess whether our online closed community heart care support group and information resource could sustain changes in health behaviour after the moderators withdrew their support. Heart patients (n = 108) living in a deprived area of Greater Manchester were recruited from general practitioners' coronary heart disease registries. The sample for this randomized controlled trial was divided in half at random where half of the participants received password-protected access to our health portal and the other half did not. At 6 months follow-up (based on the moderated phase), there was a significant difference between the experimental group and the controls in terms of self-reported diet (eating bad foods less often). This change in behaviour was not sustained during the 3-month unmoderated phase. During this unmoderated phase of the intervention, the experimental group had significantly more health care visits compared with the controls. There was no significant difference between the two phases for either group in terms of exercise, smoking or social support. This study offers insight into the potential implications for health changes of moderating arrangements for online health communities.