Obesity in childhood, which is associated with cardiovascular risk factors such as hypertension, is on the increase. Countermeasures are necessary. In this paper, we present the baseline and final data from the StEP TWO programme, a prospective study to prevent overweight and obesity in primary schools. Methods: We recorded and calculated, from 1689 children, anthropometric data, including analyses of bioelectric impedance, waist and hip circumferences, body mass index and its standard deviation, and the ratio of waist to hip. Blood pressure was measured after 5 minutes at rest. From the three schools involved in a programme of intervention, 121 children were invited to take part, and 40 (33.1 per cent) completed the programme. The effect was compared with 155 overweight and obese children identified at the 4 control schools. Results: 830 (49.5 per cent) boys and 848 girls (50.5 per cent) took part. Their mean age was 8.2 plus or minus 1.3 years, their height was 1.31 plus or minus 0.09 metres, they weighed 30.0 plus or minus 8.2 kilograms, and their mean index of body mass was 17.1 plus or minus 2.9 kilograms per metre squared. Of the children, 7.3 per cent were obese, 10.4 per cent were overweight, 75.7 per cent had normal weights, and 6.6 per cent were underweight. Resting hypertension was observed in 2.3 per cent of the children. Increased blood pressure was associated with a higher body weight, body mass index, standard deviation score for body mass index, and waist and hip circumferences (each p < 0.001), but not with the ratio of waist to hip. Hypertension at rest was also found in 11.0 per cent of obese children, 4.4 per cent of those who were overweight, 1.2 per cent of those with normal weight, and 1.0 per cent of underweight children (p < 0.001). After the intervention, the increase of the body mass index tended to be lower in those in whom we had intervened (p = 0.069), and in these the decrease of the standard deviation score for body mass index was significantly higher (p = 0.028). Systolic blood pressure was reduced by about 10 millimetres of mercury in those in whom we had intervened (p = 0.002), while there were no changes in the control group. Diastolic blood pressure was lowered by 3 millimetres of mercury, but this was not significant. Conclusion: Obese children had the highest values for systolic and diastolic blood pressure. Increased levels of blood pressure are associated with other parameters of obesity, such as the circumference of the waist and hip. Early preventive measurements in childhood are necessary, and appropriate intervention appears to be effective.