A controlled trial was conducted in two Belgian towns to investigate the feasibility and effects of a reduction in salt consumption at the community level. The low-sodium intervention in one town was mainly directed at women and implemented through mass media techniques, while the control town was merely observed. Cross-sectional random sampling at baseline and 5 years later was employed, the participation rate being similar (67%) in the two towns. During the study a total of 2211 subjects were examined. In adult women (greater than or equal to 20 years) in the intervention town the 24-h urinary excretion of sodium (UVNa) decreased by 25 mmol/24 h (P less than 0.001) and this reduction differed (P = 0.01) from the concurrent trend in UVNa in the control town (+8 mmol/24 h). However, both systolic (SBP, -7.5 versus -7.9 mmHg) and diastolic (DBP, -2.3 versus -3.0 mmHg) pressures declined to a similar extent in the women from the two towns. In adult men in the intervention town, decreases were observed in UVNa (-12 mmol/24 h) and in SBP (-5.6 mmHg) and DBP (-2.4 mmHg), but these trends were not significantly different from the concurrent changes in the control town (-14 mmol/24 h, -4.9 and +0.2 mmHg, respectively).(ABSTRACT TRUNCATED AT 250 WORDS)