Study designs of evaluations included in the review
Randomised controlled trials (RCTs), non-randomised controlled before-and-after studies, and non-randomised controlled trials and interrupted time series analyses. Studies ranged from an intervention of one session to regular meetings for at least a year.
In the 5 studies including women of childbearing age, all studies were community-based. In the 4 studies including pregnant women, 2 were community-based and 2 studies were clinic-based.
Specific interventions included in the review
Interventions to promote healthy eating including health education, counselling, changes in environment and changes in policy. In the 5 studies including women of childbearing age, 2 studies were predominantly educational, 2 studies involved an exercise programme and 1 study offered empowerment and support to new mothers. In the 4 studies of pregnant women, all of the studies had an educational component and in three there was also counselling and/or various levels of psychosocial support.
Healthy eating was defined as any target outcome broadly in line with the National Advisory Committee on Nutrition Education (NACNE) 1983 recommendations (see Other Publications of Related Interest no.1).
Participants included in the review
Pregnant women and women of childbearing age. Childbearing-age participants were aged 15-45 or thereabouts. Participants were free-living (i.e. not living in an institution, in prison or a hospital).
Outcomes assessed in the review
Changes in knowledge and attitudes about healthy eating and dietary intake (particularly fat intake). Specifically:
1. The extent of the reported change in intake of key dietary indicators in participants (experimental and control).
2. Quantitative differences in the size of changes in intake of key dietary indicators in the two types of participants.
Interventions of weight management in overweight participants and studies reporting the effect of supplementation with vitamins and other nutrients were excluded from this review.
How were decisions on the relevance of primary studies made?
Two independent reviewers selected the articles for inclusion. One of the reviewers assessed all of the studies and three other reviewers each assessed one-third of the retrieved articles. Disagreement was resolved by a review by a third reviewer and a final decision made after discussion.