As a test of the social support hypothesis, highly anxious primiparous mothers were assigned in the post-natal stage to either a professional intervention, a lay intervention or to a control group. It was hypothesised that those receiving an active intervention (be it lay or professional assistance) would become less anxious as a consequence of a central therapeutic ingredient – social support. Improvement was assessed by measuring state anxiety levels at baseline and at 12 months, while the degree to which therapists were incorporated into the social network was assessed by the Interview Schedule for Social Interaction (ISSI), given at baseline and at 12 months. While we established that anxiety levels were significantly lowered in those receiving the professional intervention and moderately (but not significantly) lowered in those receiving the lay intervention, ISSI scores for the separate groups appeared stable over the study. Reasons are considered why ISSI scores remained unchanged while intervention groups showed a reduction in anxiety levels.